EFFICIENCY

Query Distillation: BERT-based Distillation for Ensemble Ranking (2020.coling-industry)

Challenge:	Multi-agent systems powered by large language models still face challenges . tutorial focuses on three core components to build effective and efficient systems .
Approach:	This tutorial introduces recent advances in building effective and efficient multi-agent LLM systems . it focuses on three core components: model distillation, dynamic routing, memory- and compute efficient serving .
Outcome:	This tutorial introduces state-of-the-art techniques for building efficient and efficient multi-agent LLM systems . it covers coordination and communication among agents, crucial for collective performance .

Copied to clipboard

Wangshu Zhang, Junhong Liu, Zujie Wen, Yafang Wang, Gerard de Melo

Intuitive Fine-Tuning: Towards Simplifying Alignment into a Single Process (2025.acl-long)

Challenge:	Recent years have witnessed substantial progress in the development of neural ranking networks, but an increasingly heavy computational burden due to growing numbers of parameters and the adoption of model ensembles.
Approach:	They propose a two-stage distillation method that allows a smaller student model to be trained while benefiting from the better performance of the teacher model.
Outcome:	The proposed method shows higher-quality rankings compared to the teacher model.

Copied to clipboard

Ermo Hua, Biqing Qi, Kaiyan Zhang, Kai Tian, Xingtai Lv, Ning Ding, Bowen Zhou

AutoFigure-Edit: Generating Editable Scientific Illustrations via Reference-Guided Styling (2026.acl-demo)

Challenge:	Supervised Fine-Tuning (SFT) and Preference Optimization (PO) are key processes for aligning Language Models with human preferences post pre-training.
Approach:	They propose to combine Supervised Fine-Tuning and Preference Optimization (PO) with two sub-processes defined at token level within the Markov Decision Process (MDP)
Outcome:	The proposed process performs comparably or even superiorly to SFT and some typical PO methods across several tasks, particularly those requires generation, reasoning, and fact-following abilities.

Copied to clipboard

Zhen Lin, Qiujie Xie, Minjun Zhu, Shichen Li, QiYao Sun, Enhao Gu, Yiran Ding, Ke Sun, Fang Guo, Panzhong Lu, Zhiyuan Ning, Yixuan Weng, Yue Zhang

EffEval: A Comprehensive Evaluation of Efficiency for MT Evaluation Metrics (2023.findings-emnlp)

Challenge:	Existing automated systems for scientific illustrations are limited in editability, stylistic controllability, and efficiency.
Approach:	They propose an end-to-end system that generates fully editable scientific illustrations from long-form scientific text while enabling flexible style adaptation through user-provided reference images.
Outcome:	The proposed system generates fully editable scientific illustrations from long-form scientific texts while enabling flexible style adaptation through user-provided reference images.

Copied to clipboard

Daniil Larionov, Jens Grünwald, Christoph Leiter, Steffen Eger

Text Generation

Challenge:	a recent surge of interest in developing evaluation metrics based on pretrained large language models (LLMs) can better cope with lexical variation.
Approach:	They propose to replace computation-intensive transformers with lighter alternatives and employ linear and quadratic approximations for alignment algorithms on top of LLM representations.
Outcome:	The proposed approach replaces computation-intensive transformers with lighter alternatives and employs linear and quadratic approximations for alignment algorithms on top of LLM representations.

A Confidence-based Acquisition Model for Self-supervised Active Learning and Label Correction (2025.tacl-1)

Copied to clipboard

Carel van Niekerk, Christian Geishauser, Michael Heck, Shutong Feng, Hsien-chin Lin, Nurul Lubis, Benjamin Ruppik, Renato Vukovic, Milica Gašić

COMBO: State-of-the-Art Morphosyntactic Analysis (2021.emnlp-demo)

Challenge:	Existing approaches to training deep neural networks require large amounts of meticulously annotated data.
Approach:	They propose a pool-based active learning framework that requires expert annotators to label only a fraction of a sequence and facilitates self-supervision for the remainder of the sequence.
Outcome:	The proposed model outperforms baselines on dialogue belief tracking tasks.

Copied to clipboard

Mateusz Klimaszewski, Alina Wróblewska

SELENE: Selective and Evidence-Weighted LLM Debating for Efficient and Reliable Reasoning (2026.eacl-industry)

Challenge:	COMBO is an end-to-end NLP system for accurate part-of-speech tagging, morphological analysis, and (enhanced) dependency parsing.
Approach:	They propose a fully neural NLP system for accurate part-of-speech tagging, morphological analysis, lemmatisation, and (enhanced) dependency parsing.
Outcome:	The proposed system predicts categorical morphosyntactic features whilst also exposes their vector representations, extracted from hidden layers.

Copied to clipboard

Akshay Verma, Swapnil Gupta, Deepak Gupta, Prateek Sircar, Siddharth Pillai

Sliding Windows Are Not the End: Exploring Full Ranking with Long-Context Large Language Models (2025.acl-long)

Challenge:	Existing multi-agent debate frameworks are computationally expensive and prone to degradation under pro-longed debates due to redundant exchanges and unstable judging.
Approach:	They propose a framework that unifies Selective Debate Initiation (SDI) with Evidence Weighted Self-Consistency (EWSC) for adaptive, debate-on-demand reasoning.
Outcome:	Evaluated on BoolQ, CosmosQA, and an internal QnA benchmark, the proposed framework achieves higher factual robustness and efficiency.

Copied to clipboard

Wenhan Liu, Xinyu Ma, Yutao Zhu, Ziliang Zhao, Shuaiqiang Wang, Dawei Yin, Zhicheng Dou

BERxiT: Early Exiting for BERT with Better Fine-Tuning and Extension to Regression (2021.eacl-main)

Challenge:	Existing methods for listwise passage ranking use sliding window approach, which is inefficient as it requires repetitive and serialized processing.
Approach:	They propose a listwise label construction approach and importance-aware learning objective for full ranking.
Outcome:	The proposed method outperforms existing methods in listwise ranking tasks.

Copied to clipboard

Ji Xin, Raphael Tang, Yaoliang Yu, Jimmy Lin

Match, Compare, or Select? An Investigation of Large Language Models for Entity Matching (2025.coling-main)

Challenge:	Existing methods to make exiting decisions are limited to classification tasks . large-scale pre-trained language models such as BERT have brought performance gain but at the cost of heavy computational burden.
Approach:	They propose a fine-tuning strategy and a learning-to-exit module to accelerate BERT inference . they propose to make trade-offs between model quality and efficiency by early exiting .
Outcome:	The proposed approach improves early exiting for BERT, with better trade-offs . it can be combined with other acceleration methods, and the proposed strategy can be applied to regression tasks.

Copied to clipboard

Tianshu Wang, Xiaoyang Chen, Hongyu Lin, Xuanang Chen, Xianpei Han, Le Sun, Hao Wang, Zhenyu Zeng

Thesis Proposal: Efficient KV Cache Reuse for Multi-Document Retrieval-Augmented Generation (2026.eacl-srw)

Challenge:	Entity matching (EM) is a critical step in entity resolution (ER).
Approach:	They propose a method that incorporates record interactions from different perspectives.
Outcome:	The proposed framework improves on 8 ER datasets and 10 LLMs and achieves higher efficiency and effectiveness.

Copied to clipboard

Zhipeng Zhang, Dmitry Ilvovsky

SAMP: A Model Inference Toolkit of Post-Training Quantization for Text Processing via Self-Adaptive Mixed-Precision (2023.emnlp-industry)

Challenge:	Retrieval-Augmented Generation (RAG) systems face efficiency bottlenecks in prefill due to attention mechanism, and traditional KV cache only accelerates decoding.
Approach:	They propose a multi-document KV cache reuse framework for multi-doc RAG workloads . they propose to resolve position and context misalignment while eliminating document-specific quadratic complexity in prefill.
Outcome:	The proposed framework solves position and context misalignment issues while eliminating document-specific quadratic complexity in prefill.

Copied to clipboard

Rong Tian, Zijing Zhao, Weijie Liu, Haoyan Liu, Weiquan Mao, Zhe Zhao, Kan Zhou

Challenge:	Existing methods for quantization of models are too complicated and can cause performance damage.
Approach:	They propose a self-adaptive mixed-precision (SAMP) toolkit to automatically control quantization rate by a mixed-presence architecture to balance model accuracy and efficiency.
Outcome:	The proposed toolkit has a higher speedup than PyTorch and FasterTransformer while ensuring the required accuracy.

How to Make LMs Strong Node Classifiers? (2026.findings-eacl)

Copied to clipboard

Zhe Xu, Kaveh Hassani, Si Zhang, Hanqing Zeng, Michihiro Yasunaga, Limei Wang, Dongqi Fu, Ning Yao, Bo Long, Hanghang Tong

Improving Privacy Guarantee and Efficiency of Latent Dirichlet Allocation Model Training Under Differential Privacy (2021.findings-emnlp)

Challenge:	Language Models (LMs) are increasingly challenging the dominance of domain-specific models, such as Graph Neural Networks (GNNs) and Graph Transformers (GTs).
Approach:	They propose a novel approach that empowers off-the-shelf LMs to achieve performance comparable to state-of-the art (SOTA) GNNs on node classification tasks without requiring any architectural modifications.
Outcome:	The proposed approach outperforms existing GNNs on node classification tasks and is open-source upon publication.

Copied to clipboard

Tao Huang, Hong Chen

Accelerating BERT Inference for Sequence Labeling via Early-Exit (2021.acl-long)

Challenge:	Latent Dirichlet allocation (LDA) is a widely used topic model to discover the latent semantic of text data.
Approach:	They propose to combine a subsampling method with CGS to improve efficiency while amplifying privacy by using a novel metric, the efficiency–privacy function.
Outcome:	The proposed algorithm improves efficiency while amplifying privacy while subsampling in CGS increases efficiency while preserving privacy.

Copied to clipboard

Xiaonan Li, Yunfan Shao, Tianxiang Sun, Hang Yan, Xipeng Qiu, Xuanjing Huang

BOLT: Fast Energy-based Controlled Text Generation with Tunable Biases (2023.acl-short)

Challenge:	Existing early-exit mechanisms are designed for sequence-level tasks, rather than sequence labeling.
Approach:	They propose to extend sentence-level early-exit to accelerate inference of PTMs . they propose a token-level mechanism that allows partial tokens to exit early at different layers .
Outcome:	The proposed approach can save up to 66%75% inference cost with minimal performance degradation.

Copied to clipboard

Xin Liu, Muhammad Khalifa, Lu Wang

Sentiment Analysis

Challenge:	Energy-based models (EBMs) have gained popularity for controlled text generation due to their high applicability to a wide range of constraints.
Approach:	They propose a language model with tunable biases to adjust the language model’s output logits.
Outcome:	The proposed model maintains the generator’s autoregressive nature to assert a strong control on token-wise conditional dependencies and overall fluency, and converges faster.

Mimic and Conquer: Heterogeneous Tree Structure Distillation for Syntactic NLP (2020.findings-emnlp)

Copied to clipboard

Hao Fei, Yafeng Ren, Donghong Ji

Gentopia.AI: A Collaborative Platform for Tool-Augmented LLMs (2023.emnlp-demo)

Challenge:	Existing work on integrating syntactic information into neural networks uses a single tree, such as a constituency or a dependency tree.
Approach:	They propose a method to integrate heterogeneous structure knowledge into a unified sequential LSTM encoder.
Outcome:	The proposed method outperforms tree encoders on four syntax-dependent tasks and is efficient and accurate.

Copied to clipboard

Binfeng Xu, Xukun Liu, Hua Shen, Zeyu Han, Yuhan Li, Murong Yue, Zhiyuan Peng, Yuchen Liu, Ziyu Yao, Dongkuan Xu

Morpheme Matters: Morpheme-Based Subword Tokenization for Korean Language Models (2026.eacl-short)

Challenge:	Existing frameworks for Augmented Language Models lack flexibility, democratization, and holistic evaluation.
Approach:	They propose a lightweight and extensible framework for Augmented Language Models called Gentopia.
Outcome:	The proposed framework integrates language models, task formats, prompting modules, and plugins into a unified paradigm.

Copied to clipboard

DongHyeok Lee, Jeongyeon Park, Kyungbeen Cho, Jae Sung Lee

Advancing the Robustness of Large Language Models through Self-Denoised Smoothing (2024.naacl-short)

Challenge:	Existing tokenizers rely on frequency-based segmentation to represent words . this often leads to inefficient token representations and oversegmentation .
Approach:	They propose a tokenization method that emphasizes the importance of Korean morphological structures in eojeol.
Outcome:	The proposed method outperforms existing tokenizers on Korean benchmark tasks and produces significantly fewer tokens per input sequence.

Copied to clipboard

Jiabao Ji, Bairu Hou, Zhen Zhang, Guanhua Zhang, Wenqi Fan, Qing Li, Yang Zhang, Gaowen Liu, Sijia Liu, Shiyu Chang

Robustness in NLP

Challenge:	Existing adversarial attacks can cause LLMs to make wrong predictions on downstream tasks or generate harmful content misaligned with human values.
Approach:	They propose to use randomized smoothing to add noise to the input and then make predictions based on these denoised versions.
Outcome:	The proposed method surpasses existing methods in both empirical and certified robustness in defending against adversarial perturbations for both downstream tasks and human alignments (i.e., jailbreak attacks).

UltraEval: A Lightweight Platform for Flexible and Comprehensive Evaluation for LLMs (2024.acl-demos)

Copied to clipboard

Chaoqun He, Renjie Luo, Shengding Hu, Ranchi Zhao, Jie Zhou, Hanghao Wu, Jiajie Zhang, Xu Han, Zhiyuan Liu, Maosong Sun

UniICL: An Efficient ICL Framework Unifying Compression, Selection, and Generation (2025.acl-long)

Challenge:	Existing evaluation platforms are complex and poorly modularized, hindering seamless incorporation into researcher’s workflows.
Approach:	They propose a lightweight evaluation framework characterized by lightweight, comprehensiveness, modularity, and efficiency that integrates models, data, and metrics into a unified evaluation workflow.
Outcome:	The proposed evaluation framework is lightweight, comprehensive, modular, and efficient.

Copied to clipboard

Jun Gao, Qi Lv, Zili Wang, Tianxiang Wu, Ziqiang Cao, Wenjie Li

Autoregressive Entity Generation for End-to-End Task-Oriented Dialog (2022.coling-1)

Challenge:	Existing methods to improve reasoning abilities of Large Language Models (LLMs) have limitations due to excessive growth in context length, causing large hardware burden.
Approach:	They propose a novel Unified ICL framework that unifies demonstration compression, demonstration selection, and final response generation.
Outcome:	The proposed framework unifies demonstration compression, demonstration selection, and final response generation.

Copied to clipboard

Guanhuan Huang, Xiaojun Quan, Qifan Wang

Soft Self-Consistency Improves Language Models Agents (2024.acl-short)

Challenge:	Task-oriented dialog systems require external knowledge base to generate a response . current systems require scanning the KB at each turn, which is inefficient when the kb scales up .
Approach:	They propose to generate entity autoregressively before leveraging it to guide response generation.
Outcome:	Experiments on MultiWOZ 2.1 single and CAMREST show that the proposed system generates more high-quality and entity-consistent responses in an end-to-end manner.

Copied to clipboard

Han Wang, Archiki Prasad, Elias Stengel-Eskin, Mohit Bansal

MoFE: Mixture of Frozen Experts Architecture (2025.naacl-industry)

Challenge:	Current “sample and select” methods rely on majority voting to score answers . however, when tasks have many distinct and valid answers, selection by voting requires a large number of samples.
Approach:	They introduce a method that replaces SC's discontinuous scoring with a continuous score computed from model likelihoods to increase selection even when actions are sparsely distributed.
Outcome:	The proposed method improves performance and efficiency on long-horizon interactive tasks by replacing SC’s discontinuous scoring with a continuous score computed from model likelihoods.

Copied to clipboard

Jean Seo, Jaeyoon Kim, Hyopil Shin

Dissecting GraphRAG: A Modular Analysis of Knowledge Structuring for Factoid Question Answering (2026.tacl-1)

Challenge:	Large Language Models (LLMs) are characterized by their immense size, often consisting of at least one billion parameters.
Approach:	They propose a mixture of Frozen Experts architecture that integrates PEFT and MoE to enhance both training efficiency and model scalability.
Outcome:	The proposed architecture outperforms other methods while achieving the highest efficiency.

Copied to clipboard

Noriki Nishida, Rumana Ferdous Munne, Shanshan Liu, Narumi Tokunaga, Yuki Yamagata, Fei Cheng, Kouji Kozaki, Yuji Matsumoto

ScaleBox: Enabling High-Fidelity and Scalable Code Verification for Large Language Models (2026.acl-demo)

Challenge:	GraphRAG integrates structured knowledge graphs into question answering . high-quality triple extraction is critical, but lacks granularity and topical coherence . large language models suffer from inherent limitations in their internalized knowledge .
Approach:	They evaluate module-level design choices in GraphRAG for retrieval-augmented generation . they find that triple extraction is critical for accurate and comprehensive retrieval .
Outcome:	The proposed framework outperforms other retrieval-augmented generation frameworks in accuracy and efficiency.

Copied to clipboard

Jiasheng Zheng, Xin Zheng, Boxi Cao, Pengbo Wang, Zhengzhao Ma, Qiming Zhu, Jiazhen Jiang, Yaojie Lu, Hongyu Lin, Xianpei Han, Le Sun

FastAdaSP: Multitask-Adapted Efficient Inference for Large Speech Language Model (2024.emnlp-industry)

Challenge:	Existing code sandboxes fail to provide accurate verification and efficiency under high-concurrency workloads.
Approach:	They propose a high-fidelity code verification system that provides sandbox feedback for RL training and evaluation.
Outcome:	The proposed system outperforms heuristic-matching baselines on LiveCodeBench and training stability on high-concurrency workloads.

Copied to clipboard

Yichen Lu, Jiaqi Song, Chao-Han Yang, Shinji Watanabe

CODI: Compressing Chain-of-Thought into Continuous Space via Self-Distillation (2025.emnlp-main)

Challenge:	Unlike other modalities, speech has unique temporal dependencies, making efficient inference methods unexplored.
Approach:	They propose a weighted token merging framework specifically designed for speech-related tasks to improve the trade-off between efficiency and performance.
Outcome:	The proposed method achieves state-of-the-art efficiency-performance trade-off on speech-related tasks.

Copied to clipboard

Zhenyi Shen, Hanqi Yan, Linhai Zhang, Zhanghao Hu, Yali Du, Yulan He

GNN-encoder: Learning a Dual-encoder Architecture via Graph Neural Networks for Dense Passage Retrieval (2022.findings-emnlp)

Challenge:	Prior implicit CoT methods have underperformed in terms of efficiency and robustness by relying on natural language tokens for reasoning.
Approach:	They propose a training framework that compresses natural language CoT into continuous space by aligning hidden states of a designated token.
Outcome:	The proposed framework outperforms the existing state-of-the-art in 3.1x compression rate and 28.2% accuracy on GSM8k scale.

Copied to clipboard

Jiduan Liu, Jiahao Liu, Yang Yang, Jingang Wang, Wei Wu, Dongyan Zhao, Rui Yan

Debate-Feedback: A Multi-Agent Framework for Efficient Legal Judgment Prediction (2025.naacl-short)

Challenge:	Existing approaches to perform large-scale query-passage retrieval are term-based, but they lose interaction between query-pastage pairs.
Approach:	They propose to fuse query (passage) information into query representations via graph neural networks that are constructed by queries and their top retrieved passages.
Outcome:	The proposed model outperforms existing models on MSMARCO, Natural Questions and TriviaQA datasets and achieves the new state-of-the-art on these datasets.

Copied to clipboard

Xi Chen, Mao Mao, Shuo Li, Haotian Shangguan

BPP-Search: Enhancing Tree of Thought Reasoning for Mathematical Modeling Problem Solving (2025.acl-long)

Challenge:	Comparative experiments show that our model outperforms several general-purpose and domain-specific legal models.
Approach:	They propose a legal judgment prediction model that integrates LLMs with argumentative reasoning techniques to simulate the debate phase of real courtroom trials.
Outcome:	The proposed model outperforms several general-purpose and domain-specific legal models and offers a dynamic reasoning process.

Copied to clipboard

Teng Wang, Wing Yin Yu, Zhenqi He, Zehua Liu, HaileiGong HaileiGong, Han Wu, Xiongwei Han, Wei Shi, Ruifeng She, Fangzhou Zhu, Tao Zhong

Stop Guessing When to Stop Testing: Efficient Model Evaluation with Just Enough Data (2026.findings-acl)

Challenge:	Existing datasets in operations research domain lack detailed annotations of the modeling process, focusing only on objective values.
Approach:	They propose an annotation-based tree-of-thought tree-based reasoning algorithm that integrates reinforcement learning into a tree- of-though.
Outcome:	The proposed algorithm outperforms state-of-the-art methods on StructuredOR, NL4OPT, and MAMO-ComplexLP datasets.

Copied to clipboard

Ofir Arviv, Kristjan Greenewald, Yotam Perlitz, Hadar Mulian, Michal Shmueli-Scheuer, Leshem Choshen

Enabling Discriminative Reasoning in LLMs for Legal Judgment Prediction (2024.findings-emnlp)

Challenge:	Current evaluation practices, typically employing fixed-size benchmarks, are inherently wasteful, continuing to the predetermined sample size even when the CI reaches 2.5, saving 80% of the evaluation cost.
Approach:	They propose an adaptive evaluation framework that combines sequential testing with stopping criteria tailored to common evaluation needs such as diminishing returns detection and minimum detectable effect size.
Outcome:	The proposed framework reduces computational cost and reliability while maintaining statistical significance.

Copied to clipboard

Chenlong Deng, Kelong Mao, Yuyao Zhang, Zhicheng Dou

DOSE: Data Selection for Multi-Modal LLMs via Off-the-Shelf Models (2026.findings-acl)

Challenge:	Existing large language models (LLMs) underperform in legal judgment prediction due to challenges in understanding case facts and distinguishing between similar charges.
Approach:	They propose a framework that allows LLMs to discriminate among charges and a judicial reasoning framework to improve their models for effective legal judgment prediction.
Outcome:	The proposed framework improves accuracy and efficiency when dealing with complex and confusing charges.

Copied to clipboard

Biao Wu, Yiwu Zhong, Meng Fang, Ling Chen

Self-Edit: Fault-Aware Code Editor for Code Generation (2023.acl-long)

Challenge:	Existing data filtering methods are expensive because they are trained on the same data they are meant to screen.
Approach:	They propose to use off-the-shelf pretrained models that have never seen the target data to select training samples for larger and stronger multimodal models without task-specific training.
Outcome:	The proposed method can achieve comparable or even better results than those trained on the full dataset in standard VQA and math benchmarks.

Copied to clipboard

Kechi Zhang, Zhuo Li, Jia Li, Ge Li, Zhi Jin

SPARTA: Efficient Open-Domain Question Answering via Sparse Transformer Matching Retrieval (2021.naacl-main)

Challenge:	Existing Large language models (LLMs) have low pass rates and accuracy on competitive programming tasks.
Approach:	They propose a generate-and-edit approach that uses execution results of generated code from LLMs to improve code quality on competitive programming tasks.
Outcome:	The proposed method improves pass@1 by 89% on APPS-dev, 31% on apps-test, and 48% on HumanEval over nine popular code generation LLMs with parameter sizes ranging from 110M to 175B.

Copied to clipboard

Tiancheng Zhao, Xiaopeng Lu, Kyusong Lee

Efficient Low-Resource Language Models Using Tokenizer Transfer (2026.eacl-srw)

Challenge:	SPARTA is a novel neural retrieval method for open-domain question answering . it learns a sparse representation that can be efficiently implemented as an Inverted Index .
Approach:	They propose a method that learns a sparse representation that can be implemented as an Inverted Index.
Outcome:	The proposed method achieves state-of-the-art results on 4 open-domain question answering tasks and 11 retrieval question answering (ReQA) tasks.

Copied to clipboard

Gustaf Gren, Murathan Kurfali

A Versatile Adaptive Curriculum Learning Framework for Task-oriented Dialogue Policy Learning (2022.findings-naacl)

Challenge:	Tokenizer transfer allows training a model for low-resource languages without full retraining . a study of pre-trained tokenizers shows that they are more efficient than traditional training methods.
Approach:	They evaluate tokenizer transfer on models trained on language-specific corpora, Orthogonal Mapping Pursuit and Fast Vocabulary Transfer.
Outcome:	The proposed model adapts to a pre-trained model without full retraining and improves cross-lingual applicability.

Copied to clipboard

Yang Zhao, Hua Qin, Wang Zhenyu, Changxi Zhu, Shihan Wang

Multifaceted Evaluation of Audio-Visual Capability for MLLMs: Effectiveness, Efficiency, Generalizability and Robustness (2025.findings-emnlp)

Challenge:	Existing training paradigms for dialogue policy learning with brute-force random sampling are expensive and lack reliable evaluation of difficulty scores.
Approach:	They propose a flexible adaptive curriculum learning framework that integrates curriculum learning with a generic global curriculum.
Outcome:	The proposed framework improves learning performance and efficiency on three public dialogue datasets.

Copied to clipboard

Yusheng Zhao, Xiao Luo, Junyu Luo, Weizhi Zhang, Zhiping Xiao, Wei Ju, Philip S. Yu, Ming Zhang

Stateful Memory-Augmented Transformers for Efficient Dialogue Modeling (2024.findings-eacl)

Challenge:	Multi-modal large language models have been used for processing and understanding information from diverse modalities.
Approach:	They propose to evaluate the audio-visual capabilities of multi-modal large language models . they focus on effectiveness, efficiency, generalizability, and robustness .
Outcome:	The proposed models exhibit strong zero-shot and few-shot generalization abilities . their success relies heavily on the vision modality, which impairs performance when visual input is corrupted or missing.

Copied to clipboard

Qingyang Wu, Zhou Yu

Semantic Outlier Removal with Embedding Models and LLMs (2025.acl-industry)

Challenge:	Existing Transformers models are computationally expensive for long context inputs.
Approach:	They propose a transformer that can interchange information between memory states and context . they evaluate the efficiency of their model on three dialogue datasets and two language datasets .
Outcome:	The proposed model is compatible with existing transformer models and can preserve dialogue history information.

Copied to clipboard

Eren Akbiyik, João F. M. De Almeida, Rik Melis, Ritu Sriram, Viviana Petrescu, Vilhjálmur Vilhjálmsson

LECTURE4ALL: A Lightweight Approach to Precise Timestamp Detection in Online Lecture Videos (2025.acl-demo)

Challenge:	Modern text processing pipelines require robust methods to remove extraneous content while preserving a document’s core message.
Approach:	They propose a method that leverages multilingual sentence embeddings and approximate nearest-neighbor search to identify and excise unwanted text segments.
Outcome:	Experiments on HTML datasets show that SORE outperforms structural methods and yields high precision in diverse scenarios.

Copied to clipboard

Viktoria Wrobel, Simon Kazemi, Frank Hammerschmidt, Torben Hannemann, Gregor Stange, Seid Muhie Yimam, Robert Geislinger

Visual Data in NLP

Challenge:	Lecture2Go provides a vast collection of recorded lectures, but locating specific content within videos can be time-consuming.
Approach:	They present an open-source web application to improve the search experience of educational video platforms.
Outcome:	The proposed solution improves the search experience of educational video platforms.

Scaling Graph-Based Dependency Parsing with Arc Vectorization and Attention-Based Refinement (2025.naacl-short)

Copied to clipboard

Nicolas Floquet, Joseph Le Roux, Nadi Tomeh, Thierry Charnois

Pretrained Bidirectional Distillation for Machine Translation (2023.acl-long)

Challenge:	Existing graph-based dependency parsers use a standard two-pipeline approach that only scores arcs and labels .
Approach:	They propose a graph-based dependency parsing architecture that explicitly constructs vectors from which both arcs and labels are scored.
Outcome:	The proposed model outperforms state-of-the-art models on PTB and UD in accuracy and efficiency.

Copied to clipboard

Yimeng Zhuang, Mei Tu

Reinforcement Learning for Self-Improving Agent with Skill Library (2026.acl-long)

Challenge:	Existing studies have focused on language knowledge transfer from pretrained models to neural machine translation models.
Approach:	They propose to use masked language pretraining to efficiently transfer bidirectional language knowledge to NMT models.
Outcome:	The proposed method can significantly improve machine translation performance and achieve competitive or even better results than previous methods.

Copied to clipboard

Jiongxiao Wang, Qiaojing Yan, Yawei Wang, Yijun Tian, Soumya Smruti Mishra, Zhichao Xu, Megha Gandhi, Panpan Xu, Lin Lee Cheong

Challenge:	Large Language Model (LLM)-based agents have demonstrated remarkable capabilities in complex reasoning and multi-turn interactions but struggle to continuously improve and adapt when deployed in new environments.
Approach:	They propose a Reinforcement Learning-based approach to enhance agents’ self-improvement capabilities with a skill library.
Outcome:	The proposed framework achieves 8.9% higher Scenario Goal Completion when applied to supervised-finetuned model with expert experience while requiring 26% fewer interaction steps and generating 59% fewer tokens.

SQLGenie: A Practical LLM based System for Reliable and Efficient SQL Generation (2025.acl-industry)

Copied to clipboard

Pushpendu Ghosh, Aryan Jain, Promod Yenigalla

SeRTS: Self-Rewarding Tree Search for Biomedical Retrieval-Augmented Generation (2024.findings-emnlp)

Challenge:	Large Language Models (LLMs) enable natural language to SQL conversion, but generating accurate, efficient queries is challenging due to ambiguous intent, domain knowledge requirements and database constraints.
Approach:	They propose a system for reliable SQL generation that integrates Table Onboarder, SQL Generator and Feedback Augmentation.
Outcome:	The proposed system surpasses the best single-LLM baseline by 21.5% and the strongest pipeline competitor by 5.3% on public benchmarks and internal datasets.

Copied to clipboard

Minda Hu, Licheng Zong, Hongru Wang, Jingyan Zhou, Jingjing Li, Yichen Gao, Kam-Fai Wong, Yu Li, Irwin King

Client-Customized Adaptation for Parameter-Efficient Federated Learning (2023.findings-acl)

Challenge:	Existing retrieval-augmented approaches to large language models face performance limitations due to the lack of publicly available training data.
Approach:	They propose a plug-and-play LLM-based retrieval method called Self-Rewarding Tree Search based on Monte Carlo Tree Search and a self-rewarding paradigm to address these limitations.
Outcome:	The proposed method improves the performance of the BM25 retriever and surpasses the baseline of self-reflection in both efficiency and scalability.

Copied to clipboard

Yeachan Kim, Junho Kim, Wing-Lam Mok, Jun-Hyung Park, SangKeun Lee

Praat++: Multimedia Annotation System for Speech and Vocalization (2026.acl-demo)

Challenge:	Pre-trained language models have a large memory footprint and are difficult to use in federated learning (FL)
Approach:	They propose a hypernetwork-based FL framework that generates client-specific adapters by conditioning the client information.
Outcome:	The proposed framework maximizes the utility of shared model parameters while minimizing divergence caused by client heterogeneity.

Copied to clipboard

Weiran Zhang, Kenny Q. Zhu

Everything of Thoughts: Defying the Law of Penrose Triangle for Thought Generation (2024.findings-acl)

Challenge:	High-quality time-aligned annotation is fundamental to speech processing and animal vocalization research, yet precise boundary localization and consistent labeling remain challenging in collaborative settings.
Approach:	They propose a web-based multimedia annotation system for collaborative, video-informed, and AI-assisted timeline labeling of audio and video data.
Outcome:	The proposed system improves time-aligned labeling and accuracy in speech and animal vocalization annotations.

Copied to clipboard

Ruomeng Ding, Chaoyun Zhang, Lu Wang, Yong Xu, Minghua Ma, Wei Zhang, Si Qin, Saravan Rajmohan, Qingwei Lin, Dongmei Zhang

Challenge:	Recent advances in Large Language Models (LLMs) have greatly advanced problem solving in diverse domains such as mathematical reasoning and knowledge reasoning.
Approach:	They propose a thought prompting approach called 'Everything of Thoughts' it leverages pretrained reinforcement learning and Monte Carlo Tree Search to incorporate external domain knowledge and planning capability into thoughts.
Outcome:	The proposed approach outperforms existing approaches on game of 24, 8-Puzzle, and Pocket Cube.

CompileAgent: Automated Real-World Repo-Level Compilation with Tool-Integrated LLM-based Agent System (2025.acl-long)

Copied to clipboard

Li Hu, Guoqiang Chen, Xiuwei Shang, Shaoyin Cheng, Benlong Wu, LiGangyang LiGangyang, Xu Zhu, Weiming Zhang, Nenghai Yu

Challenge:	CompileAgent is the first LLM-based agent framework dedicated to repo-level compilation.
Approach:	They propose a LLM-based agent framework dedicated to repo-level compilation.
Outcome:	The proposed method significantly improves compilation success rate, ranging from 10% to 71%.

RESTful-Llama: Connecting User Queries to RESTful APIs (2024.emnlp-industry)

Copied to clipboard

Han Xu, Ruining Zhao, Jindong Wang, Haipeng Chen

A Career Interview Dialogue System using Large Language Model-based Dynamic Slot Generation (2025.coling-main)

Challenge:	Recent advances in Large Language Models (LLMs) have demonstrated exceptional performance in zero-shot learning and reasoning tasks.
Approach:	They propose a framework that transforms natural language instructions into effective RESTful API calls and a method to generate fine-tuning datasets from public API documentation.
Outcome:	The proposed framework improves performance in a 31.9% improvement in robustness and 2.33x increase in efficiency compared to existing methods.

Copied to clipboard

Ekai Hashimoto, Mikio Nakano, Takayoshi Sakurai, Shun Shiramatsu, Toshitake Komazaki, Shiho Tsuchiya

Are Large Language Models Economically Viable for Industry Deployment? (2026.acl-industry)

Challenge:	a slot-filling-based interview dialogue system is limited in the flexibility of information collection . authors propose a method that leverages large language models to generate new slots according to the flow of the dialogue .
Approach:	They propose a slot-filling dialogue system that collects information on staff careers . they incorporate abduction into the slot generation process to enable more natural conversations .
Outcome:	The proposed method improves the efficiency and quality of career interviews conducted by nursing managers.

Copied to clipboard

Abdullah Mohammad, Sushant Kumar Ray, Pushkar Arora, Rafiq Ali, Ebad Shabbir, Gautam Siddharth Kashyap, Jiechao Gao, Usman Naseem

Structured Pruning Learns Compact and Accurate Models (2022.acl-long)

Challenge:	Generative AI is increasingly deployed in healthcare, financial analytics, and conversational automation.
Approach:	They propose a framework that evaluates large language models across their full lifecycle on legacy GPUs.
Outcome:	The proposed framework evaluates LLMs across their full lifecycle on legacy GPUs.

Copied to clipboard

Mengzhou Xia, Zexuan Zhong, Danqi Chen

Fanar-Sadiq: A Multi-Agent Architecture for Grounded Islamic QA (2026.acl-industry)

Challenge:	Pre-trained language models have high costs in terms of storage, memory, and computation time.
Approach:	They propose a task-specific structured pruning method CoFi which provides highly parallelizable subnetworks and matches distillation methods in both accuracy and latency.
Outcome:	The proposed method matches the distillation methods in accuracy and latency without resorting to unlabeled data.

Copied to clipboard

Ummar Abbas, Mourad Ouzzani, Mohamed Y. Eltabakh, Omar Sinan, Gagan Bhatia, Hamdy Mubarak, Majd Hawasly, Mohammed Qusay Hashim, Kareem Mohamed Darwish, Firoj Alam

WaveFM: A High-Fidelity and Efficient Vocoder Based on Flow Matching (2025.naacl-long)

Challenge:	Large language models (LLMs) can answer religious knowledge queries fluently, but they often hallucinate and misattribute sources.
Approach:	They propose a bilingual Arabic-English Islamic QA system that uses a multi-agent, tool-augmented architecture to route Islamic queries to specialized modules.
Outcome:	The proposed system is based on a multi-agent, tool-augmented architecture and has received over 1.9M accesses in less than a year.

Copied to clipboard

Tianze Luo, Xingchen Miao, Wenbo Duan

Aegis:An Advanced LLM-Based Multi-Agent for Intelligent Functional Safety Engineering (2024.emnlp-industry)

Challenge:	Flow matching is a robust and stable approach to training diffusion models, but it can result in subpar audio quality.
Approach:	They propose a reparameterized flow matching model for mel-spectrogram conditioned speech synthesis that uses a mel prior instead of a standard Gaussian prior to minimize unnecessary transportation costs.
Outcome:	The proposed model improves sample quality and generation speed for speech vocoders while reducing transportation costs.

Copied to clipboard

Lu Shi, Bin Qi, Jiarui Luo, Yang Zhang, Zhanzhao Liang, Zhaowei Gao, Wenke Deng, Lin Sun

Query-as-context Pre-training for Dense Passage Retrieval (2023.emnlp-main)

Challenge:	Aegis is an advanced LLM-based multi-agent for intelligent functional safety engineering that can perform all phases of a vehicle's lifecycle, including design, development, production, operation, and decommissioning.
Approach:	They introduce Aegis: An Advanced LLM-Based Multi-Agent for Intelligent Functional Safety Engineering.
Outcome:	The proposed solution can perform Hazard Analysis and Risk Assessment (HARA), document Functional Safety Requirements (FSR), and plan test cases for Automatic Emergency Braking (AEB) systems.

Copied to clipboard

Xing W, Guangyuan Ma, Wanhui Qian, Zijia Lin, Songlin Hu

AWARE: Agentic Knowledge Warehousing for Contextual Intelligence (2026.findings-acl)

Challenge:	Existing methods to improve passage retrieval performance by using context-supervised pre-training are weakly correlated.
Approach:	They propose to use query-as-context pre-training to train passage-query pairs . they evaluate the pre-trained models on large-scale passage retrieval benchmarks .
Outcome:	The proposed technique improves performance on large-scale passage retrieval benchmarks and out-of-domain zero-shot benchmarks.

Copied to clipboard

Hongjin Qian, Siqi Bao, Zhao Cao, Zheng Liu

Optimizing Packing and Shuffling Strategies for Enhanced Performance in Generative Language Models (2026.acl-srw)

Challenge:	Large language models excel in information seeking tasks, but their knowledge is limited in coverage and timeliness.
Approach:	They propose an agentic knowledge warehousing framework that transforms unstructured data into minimal, task-conditioned knowledge representations consumable by LLMs.
Outcome:	Experiments on GAIA, WebWalker, and BrowseComp-Plus show improvements over baselines.

Copied to clipboard

Yanbing Chen, Ruilin Wang, Zihao Yang, Lavender Yao Jiang, Eric Karl Oermann

Probing Structured Pruning on Multilingual Pre-trained Models: Settings, Algorithms, and Efficiency (2022.acl-long)

Challenge:	Documents are typically concatenated to chunks of maximum sequence length (MSL) and shuffled in chunks (atom-size chunks).
Approach:	They propose to pack and shuff documents in chunks of tokens to prevent overfitting . they also propose to use padding to only include one document per chunk .
Outcome:	The proposed method reduces the risk of overfitting and improves generalizability.

Copied to clipboard

Yanyang Li, Fuli Luo, Runxin Xu, Songfang Huang, Fei Huang, Liwei Wang

Efficient Model Development through Fine-tuning Transfer (2025.emnlp-main)

Challenge:	Structured pruning has been extensively studied on monolingual pre-trained models . but little attention has been paid to evaluating the effectiveness of structured pruning on multilingual models.
Approach:	They investigate settings, algorithms, and efficiency of structured pruning on multilingual models . authors propose a simple approach that allows training the model once and adapting to different model sizes at inference .
Outcome:	The proposed approach allows training the model once and adapting to different model sizes at inference.

Copied to clipboard

Pin-Jie Lin, Rishab Balasubramanian, Fengyuan Liu, Nikhil Kandpal, Tu Vu

Multi-document Summarization with Maximal Marginal Relevance-guided Reinforcement Learning (2020.emnlp-main)

Challenge:	Modern large language models face a major bottleneck: each new version of a pre-trained model requires expensive and repetitive alignment.
Approach:	They propose a method that transfers fine-tuning updates across model versions . they extract the diff vector, which is the difference in parameters induced by fine-uning, from a source model and apply it to the base of a different target model.
Outcome:	The proposed method reduces training costs while maintaining model performance.

Copied to clipboard

Yuning Mao, Yanru Qu, Yiqing Xie, Xiang Ren, Jiawei Han

Optimal Neural Program Synthesis from Multimodal Specifications (2021.findings-emnlp)

Challenge:	Recent studies on single-document summarization (SDS) benefit from advances in neural sequence learning, but they produce unsatisfactory results on multi-document summary (MDS).
Approach:	They propose a neural sequence learning method that unifies advanced neural SDS methods and statistical measures used in classical MDS.
Outcome:	The proposed method achieves state-of-the-art performance on benchmark MDS datasets.

Copied to clipboard

Xi Ye, Qiaochu Chen, Isil Dillig, Greg Durrett

ArcaneQA: Dynamic Program Induction and Contextualized Encoding for Knowledge Base Question Answering (2022.coling-1)

Challenge:	Existing methods for multimodal program synthesis combine noisy signals from the user with hard constraints on the program’s behavior.
Approach:	They propose an optimal neural synthesis approach where the goal is to find a program that satisfies user-provided constraints while also maximizing the program’s score with respect to a neural model.
Outcome:	The proposed approach outperforms prior state-of-the-art methods in terms of accuracy and efficiency and finds model-optimal programs more frequently.

Copied to clipboard

Yu Gu, Yu Su

Aristotle: Mastering Logical Reasoning with A Logic-Complete Decompose-Search-Resolve Framework (2025.acl-long)

Challenge:	Existing ranking-based KBQA models struggle with flexibility in predicting complicated queries and have impractical running time.
Approach:	They propose a new generation-based question answering on knowledge bases model that addresses both large search space and ambiguities in schema linking.
Outcome:	The proposed model overcomes two intertwined challenges on popular KBQA datasets and is highly competitive and efficient.

Copied to clipboard

Jundong Xu, Hao Fei, Meng Luo, Qian Liu, Liangming Pan, William Yang Wang, Preslav Nakov, Mong-Li Lee, Wynne Hsu

Neural Storyline Extraction Model for Storyline Generation from News Articles (N18-1)

Challenge:	Existing systems fail to fully leverage the structure of logical tasks throughout the reasoning process, causing bottlenecks in efficiency and efficacy.
Approach:	They propose a logic-complete reasoning framework, Aristotle, which integrates symbolic expressions and logical rules into the entire reasoning process.
Outcome:	The proposed framework outperforms state-of-the-art reasoning frameworks in accuracy and efficiency.

Copied to clipboard

Deyu Zhou, Linsen Guo, Yulan He

Mitigating Context Interference for Reliable and Efficient Search Agents (2026.acl-long)

Challenge:	Existing approaches to storyline generation are domain dependent and cannot deal with unseen event types.
Approach:	They propose a neural network-based approach to extract structured representations and evolution patterns of storylines without using annotated data.
Outcome:	The proposed model outperforms state-of-the-art approaches on accuracy and efficiency on three news corpora and it is based on supervised models.

Copied to clipboard

Boyang Xue, Bin Wu, Shuofei Qiao, Sheng Wang, Rui Wang, Yiming Du, Hongru Wang, Jeff Z. Pan, Emine Yilmaz, Kam-Fai Wong, Aldo Lipani

SHIFT: Selected Helpful Informative Frame for Video-guided Machine Translation (2025.emnlp-main)

Challenge:	Recent research empowers Large Language Models (LLMs) as multi-turn search agents to iteratively retrieve and generate outputs until complex tasks are solved.
Approach:	They propose a distill-based context refiner to dynamically mitigate context interference . they also propose RLs that refine contexts to generate outputs .
Outcome:	The proposed refiner can mitigate context interference in multi-turn search agents.

Copied to clipboard

Boyu Guan, Chuang Han, Yining Zhang, Yupu Liang, Zhiyang Zhang, Yang Zhao, Chengqing Zong

FunnelRAG: A Coarse-to-Fine Progressive Retrieval Paradigm for RAG (2025.findings-naacl)

Challenge:	Video-guided machine translation (VMT) aims to improve translation quality by integrating contextual information from paired short video clips.
Approach:	They propose a plug-and-play framework for video-guided machine translation with multimodal large language models.
Outcome:	The proposed framework improves performance of MLLMs while reducing computational cost.

Copied to clipboard

Xinping Zhao, Yan Zhong, Zetian Sun, Xinshuo Hu, Zhenyu Liu, Dongfang Li, Baotian Hu, Min Zhang

ADaPT: Token-Level Decoupling for Efficient Large Reasoning Models (2026.findings-acl)

Challenge:	Retrieval-Augmented Generation (RAG) is widely adopted in Large Language Models, but is flat and has limitations such as a significant burden on one retriever and constant granularity limits the ceiling of retrieval performance.
Approach:	They propose a progressive retrieval paradigm with coarse-to-fine granularity for RAG, termed FunnelRAG, so as to balance effectiveness and efficiency.
Outcome:	The proposed paradigm achieves comparable retrieval performance while the time overhead is reduced by nearly 40%.

Copied to clipboard

Tingyun li, Zishang Jiang, Jinyi Han, Xinyi Wang, Sihang Jiang, Han Xia, Zhaoqian Dai, Ma Shuguang, Fei Yu, Jiaqing Liang, Yanghua Xiao

Efficient CTC Regularization via Coarse Labels for End-to-End Speech Translation (2023.eacl-main)

Challenge:	Existing efficiency-oriented methods attempt to shorten or mix reasoning strategies, yet often degrade reasoning capability.
Approach:	They propose a token-level dual-process framework that explicitly decouples efficiency and correctness signals during training.
Outcome:	The proposed framework reduces inference cost while maintaining strong reasoning ability across multiple benchmarks.

Copied to clipboard

Biao Zhang, Barry Haddow, Rico Sennrich

Speech & Audio in NLP

Challenge:	Developing techniques to support end-to-end speech translation is non-trivial because of the speech-text modality gap.
Approach:	They propose a coarse labeling approach that merges vocabulary labels via simple heuristic rules . they propose to use 256-bit truncation, division or modulo operations to regularize the encoder .
Outcome:	The proposed method can increase training efficiency while delivering better performance.

Compressing then Matching: An Efficient Pre-training Paradigm for Multimodal Embedding (2026.acl-long)

Copied to clipboard

Da Li, Yuxiao Luo, Keping Bi, Jiafeng Guo, Wei Yuan, Biao Yang, Yan Wang, Fan Yang, Tingting Gao, Guorui Zhou

WASA: A Web Application for Sequence Annotation (L18-1)

Challenge:	Recent approaches demonstrate that MLLMs can be adapted into competitive embedding models via large-scale contrastive learning.
Approach:	They propose a compressed pre-training phase which serves as a warm-up stage for contrastive learning.
Outcome:	The proposed model achieves state-of-the-art among MLLMs of comparable size on the MMEB, realizing optimization in both efficiency and effectiveness.

Copied to clipboard

Fahad AlGhamdi, Mona Diab

AdaptThink: Reasoning Models Can Learn When to Think (2025.emnlp-main)

Challenge:	a major barrier to research on CS has been the lack of large multilingual, multi-genre CS-annotated corpora.
Approach:	They propose a web-based annotation system that manages large-scale CS data annotation.
Outcome:	The proposed system can manage large-scale multilingual code switching (CS) data annotation.

Copied to clipboard

Jiajie Zhang, Nianyi Lin, Lei Hou, Ling Feng, Juanzi Li

SP3: Enhancing Structured Pruning via PCA Projection (2024.findings-acl)

Challenge:	Recent advances in large reasoning models have demonstrated remarkable capabilities in tackling complex tasks.
Approach:	They propose an algorithm to teach reasoning models to choose the optimal thinking mode based on problem difficulty.
Outcome:	The proposed algorithm reduces the average response length and improves accuracy on three math datasets.

Copied to clipboard

Yuxuan Hu, Jing Zhang, Zhe Zhao, Chen Zhao, Xiaodong Chen, Cuiping Li, Hong Chen

SEEKR: Selective Attention-Guided Knowledge Retention for Continual Learning of Large Language Models (2024.emnlp-main)

Challenge:	Structured pruning is a widely used technique for reducing the size of pre-trained language models, but current methods overlook the potential of compressing the hidden dimension d in PLMs.
Approach:	They propose a structured pruning approach that projectes features into a space defined by principal components before masking the hidden dimension d in pre-trained language models.
Outcome:	Experiments on benchmarks show that SP3 can reduce d by 70%, compress 94% of the BERTbase model, and maintain over 96% accuracy.

Copied to clipboard

Jinghan He, Haiyun Guo, Kuan Zhu, Zihan Zhao, Ming Tang, Jinqiao Wang

Language-to-Space Programming for Training-Free 3D Visual Grounding (2025.emnlp-main)

Challenge:	Existing methods fail to fully exploit the knowledge embedded in models from previous tasks . Existing techniques fail to exploit the information embedded in previous tasks, resulting in a large number of replay samples to achieve good results.
Approach:	They propose a method that uses attention weights to extract knowledge from previous tasks . they use a data replay strategy to extract the knowledge from the previous tasks.
Outcome:	The proposed method achieves comparable or even better performance with only 1/10 of replayed data used by other methods.

Copied to clipboard

Boyu Mi, Hanqing Wang, Tai Wang, Yilun Chen, Jiangmiao Pang

AutoDSPy: Automating Modular Prompt Design with Reinforcement Learning for Small and Large Language Models (2025.emnlp-industry)

Challenge:	Existing methods for 3D visual grounding have been proposed, but they are limited by the scarcity of 3D vision-language datasets and the high cost of annotations.
Approach:	They propose a method for training-free 3D visual grounding that uses LLM-generated codes to analyze 3D spatial relations among objects.
Outcome:	The proposed method achieves 52.9% accuracy on the Nr3D benchmark and significantly reduces grounding time and token costs.

Copied to clipboard

Nafew Azim, Abrar Ur Alam, Hasan Bin Omar, Abdullah Mohammad Muntasir Adnan Jami, Jawad Ibn Ahad, Muhammad Rafsan Kabir, Md. Ismail Hossain, Fuad Rahman, Mohammad Ruhul Amin, Shafin Rahman, Nabeel Mohammed

F2TEval: Human-Aligned Multi-Dimensional Evaluation for Figure-to-Text Task (2025.emnlp-main)

Challenge:	Large Language Models excel at complex reasoning tasks, yet their performance hinges on the quality of their prompts and pipeline structures.
Approach:	They propose a framework that fully automates large language models' pipeline construction using reinforcement learning.
Outcome:	Experimental results show that autoDSPy outperforms DSPy benchmarks in accuracy gains and time.

Copied to clipboard

Tan Yue, Rui Mao, Zilong Song, Zonghai Hu, Dongyan Zhao

When and How to Augment Your Input: Question Routing Helps Balance the Accuracy and Efficiency of Large Language Models (2025.findings-naacl)

Challenge:	Existing evaluation methods for Figure-to-Text tasks are limited due to the inherent ambiguity and semantic compression of figures, the generated texts suffer from factual inaccuracies, incomplete coverage, and weak logical reasoning.
Approach:	They propose a five-dimensional reference-free evaluation method aligned with expert criteria to support fine-grained evaluation.
Outcome:	The proposed method outperforms Gemini-2.0 and Claude-3.5 with only 0.9B parameters.

Copied to clipboard

Shufan Chen, He Zheng, Lei Cui

GKT: A Novel Guidance-Based Knowledge Transfer Framework For Efficient Cloud-edge Collaboration LLM Deployment (2024.findings-acl)

Challenge:	augmented generation of knowledge-based long-tail questions can be useful for large language models, but can cause significant latency.
Approach:	They propose an adaptive question routing framework that uses a query router to augment input to the right time.
Outcome:	The proposed framework surpasses existing approaches in accuracy and efficiency on benchmarks such as AmbigNQ, HotpotQA, MMLU-STEM, and PopQA.

Copied to clipboard

Yao Yao, Zuchao Li, Hai Zhao

Semantic Scaffolds for Pseudocode-to-Code Generation (2020.acl-main)

Challenge:	Existing methods of acceleration require fine-tuning of considerably large models, such as Llama-7B, posing a challenge for average users.
Approach:	They propose a Guidance-based Knowledge Transfer framework that leverages a larger LLM as a 'teacher' and a smaller 'student' model to finalize responses.
Outcome:	The proposed framework achieves a maximum accuracy improvement of 14.18%, along with a 10.72 times speed-up on GSM8K and an accuracy improvement 14.00% along with 7.73 times speed up in CSQA.

Copied to clipboard

Ruiqi Zhong, Mitchell Stern, Dan Klein

Tricking Retrievers with Influential Tokens: An Efficient Black-Box Corpus Poisoning Attack (2025.naacl-long)

Challenge:	Existing methods for program generation use lightweight structures to represent high-level semantics and syntactic composition of a program.
Approach:	They propose a method for program generation based on semantic scaffolds . they use line-level natural language pseudocode annotations to search for programs .
Outcome:	The proposed method achieves 10% improvement in top-100 accuracy over the current state-of-the-art method.

Copied to clipboard

Cheng Wang, Yiwei Wang, Yujun Cai, Bryan Hooi

Silencing the Guardrails: Inference-Time Jailbreaking via Dynamic Contextual Representation Ablation (2026.findings-acl)

Challenge:	Existing methods for crafting adversarial passages are slow and computationally expensive, requiring either access to retriever’s gradients or large computational resources.
Approach:	They propose a method that leverages two key characteristics of retrievers: insensitivity to token order and bias towards influential tokens to generate effective adversarial passages.
Outcome:	The proposed method achieves superior efficiency and scalability compared to existing methods while maintaining comparable or better attack success rates across multiple datasets.

Copied to clipboard

Wenpeng Xing, Moran Fang, Guangtai Wang, Changting Lin, Meng Han

Representation Learning

Challenge:	Existing strategies to circumvent safety constraints face significant trade-offs between effectiveness and efficiency.
Approach:	They propose a framework that allows to infer model refusal behaviors without expensive parameter updates or training.
Outcome:	The proposed framework outperforms baselines in multiple safety-aligned open-source LLMs.

Addressing Overthinking in Large Vision-Language Models via Gated Perception-Reasoning Optimization (2026.findings-acl)

Copied to clipboard

Xingjian Diao, Zheyuan Liu, Chunhui Zhang, Weiyi Wu, Keyi Kong, Lin Shi, Kaize Ding, Soroush Vosoughi, Jiang Gui

Interpretable Proof Generation via Iterative Backward Reasoning (2022.naacl-main)

Challenge:	Prior work has attempted to mitigate this issue by using adaptive reasoning strategies, but these methods overlook a fundamental bottleneck: visual perception failures.
Approach:	They propose a meta-reasoning controller that dynamically routes computation among three decision paths at each generation step.
Outcome:	The proposed method outperforms slow-thinking methods while producing shorter responses.

Copied to clipboard

Hanhao Qu, Yu Cao, Jun Gao, Liang Ding, Ruifeng Xu

An Efficient and Precise Training Data Construction Framework for Process-supervised Reward Model in Mathematical Reasoning (2025.acl-long)

Challenge:	Existing proof generation tasks require reasoning capabilities, but they usually just request for an answer without the reasoning procedure that would make it interpretable.
Approach:	They propose an iterative backward reasoning model to solve the proof generation tasks on rule-based Question Answering.
Outcome:	The proposed model improves in-domain performance and cross-domain transferability over existing models.

Copied to clipboard

Wei Sun, Qianlong Du, Fuwei Cui, Jiajun Zhang

INFORM : Information eNtropy based multi-step reasoning FOR large language Models (2023.emnlp-main)

Challenge:	Existing methods for constructing process supervision training data are costly or suffer from poor quality.
Approach:	They propose a framework called EpicPRM which annotates each intermediate reasoning step based on its quantified contribution and uses an adaptive binary search algorithm to enhance annotation precision and efficiency.
Outcome:	The proposed framework improves annotation precision and efficiency and can be used to train a high-quality training dataset with 50k annotated intermediate steps.

Copied to clipboard

Chuyue Zhou, Wangjie You, Juntao Li, Jing Ye, Kehai Chen, Min Zhang

MBA-RAG: a Bandit Approach for Adaptive Retrieval-Augmented Generation through Question Complexity (2025.coling-main)

Challenge:	Large language models (LLMs) have demonstrated exceptional performance with dedicated Chain-of-Thought (CoT) prompts.
Approach:	They propose a new method by introducing information entropy as a criteria on for CoT prompt selection.
Outcome:	The proposed model outperforms existing models on seven reasoning benchmarks using two language models.

Copied to clipboard

Xiaqiang Tang, Qiang Gao, Jian Li, Nan Du, Qi Li, Sihong Xie

Finding RELIEF: Shaping Reasoning Behavior without Reasoning Supervision via Belief Engineering (2026.findings-acl)

Challenge:	Existing RAG frameworks either indiscriminately perform retrieval or rely on rigid single-label classifiers to select retrieval methods.
Approach:	They propose a framework that dynamically selects the most suitable retrieval strategy based on query complexity.
Outcome:	The proposed framework achieves state-of-the-art results on multiple single-hop and multi-hop datasets while reducing retrieval costs.

Copied to clipboard

Chak Tou Leong, Dingwei Chen, Heming Xia, Qingyu Yin, Sunbowen Lee, Jian Wang, Wenjie Li

Efficient and Effective Prompt Tuning via Prompt Decomposition and Compressed Outer Product (2025.naacl-long)

Challenge:	Existing methods for shaping large reasoning models rely on reinforcement learning or fine-tuning with gold-standard reasoning traces. Existing techniques for behavior shaping rely only on additional reward modeling.
Approach:	They propose a framework that aligns a model's self-concept with a target belief blueprint and internalizes desired traits by fine-tuning on synthesized, self-reflective QA pairs that affirm the target belief.
Outcome:	The proposed framework outperforms behavior-supervised and preference-based models while requiring significantly lower training costs.

Copied to clipboard

Pengxiang Lan, Haoyu Xu, Enneng Yang, Yuliang Liang, Guibing Guo, Jianzhe Zhao, Xingwei Wang

LeeBERT: Learned Early Exit for BERT with cross-level optimization (2021.acl-long)

Challenge:	Existing methods for fine-tuning pre-trained language models overlook intrinsic semantic associations between soft prompt tokens, leading to high discreteness and limited interactions.
Approach:	They propose a low-parameters Prompt Tuning method which leverages prompt decomposition and compressed outer product to facilitate multiple interactions among prompt tokens.
Outcome:	Experiments on six architectures and eight datasets show that the proposed method outperforms state-of-the-art methods in performance and efficiency.

Copied to clipboard

Wei Zhu

Text is All You Need: LLM-enhanced Incremental Social Event Detection (2025.acl-long)

Challenge:	Pre-trained language models are resource exhaustive and computationally expensive for industrial scenarios.
Approach:	They propose a learning scheme to learn from each other to speed up inference . they ask each exit to learn the weights of different loss terms, instead of learning only from the last layer .
Outcome:	The proposed scheme improves state-of-the-art (SOTA) early exit methods for pre-trained models on the GLUE benchmark.

Copied to clipboard

Zitai Qiu, Congbo Ma, Jia Wu, Jian Yang

Entropy-Gated Branching for Efficient Test-Time Reasoning (2026.eacl-long)

Challenge:	Existing state-of-the-art (SOTA) SED models rely on graph neural networks (GNNs) Existing SED frameworks rely heavily on GNNs, which require complex graph construction and time-consuming training processes.
Approach:	They propose a framework that leverages the rich background knowledge of large language models to formalize and disambiguate short texts by completing abbreviations and summarizing informal expressions.
Outcome:	The proposed framework outperforms existing models on two challenging real-world datasets.

Copied to clipboard

Xianzhi Li, Ethan Callanan, Abdellah Ghassel, Xiaodan Zhu

Scalable Data Synthesis through Human-like Cognitive Imitation and Data Recombination (2025.emnlp-main)

Challenge:	Empirical results show that branching at low uncertainty points can improve reasoning capabilities of large language models . however, these methods require substantially more computational resources, causing errors in high-stakes domains .
Approach:	They propose an inference technique that selectively expands prediction sequences at points of high uncertainty.
Outcome:	Empirical results show that the proposed method improves accuracy by 22.6% over standard inference while operating 31%-75% faster across math benchmarks.

Copied to clipboard

Zhongyi Ye, Weitai Zhang, Xinyuan Zhou, Yongxin Zhu, Ninghui Rao, Enhong Chen

Rethinking Schema Linking: A Context-Aware Bidirectional Retrieval Approach for Text-to-SQL (2026.findings-eacl)

Challenge:	Large language models (LLMs) rely on massive amounts of training data, however, the quantity of empirically observed data is limited.
Approach:	They propose a data synthesis framework that mimics human cognitive behaviors by recombining and interconnecting heterogeneous data from diverse sources.
Outcome:	The proposed framework mimics human cognitive behaviors by recombining and interconnecting heterogeneous data from diverse sources thereby enhancing advanced reasoning capabilities in large language models.

Copied to clipboard

Md Mahadi Hasan Nahid, Davood Rafiei, Weiwei Zhang, Yong Zhang

GNNavi: Navigating the Information Flow in Large Language Models by Graph Neural Network (2024.findings-acl)

Challenge:	Recent methods focus on improving SQL generation but neglect retrieval of relevant schema elements.
Approach:	They propose a context-aware bidirectional schema retrieval framework that treats schema linking as a standalone problem.
Outcome:	The proposed framework improves schema recall while reducing false positives.

Copied to clipboard

Shuzhou Yuan, Ercong Nie, Michael Färber, Helmut Schmid, Hinrich Schuetze

Adaptive Feature-based Low-Rank Compression of Large Language Models via Bayesian Optimization (2024.findings-emnlp)

Challenge:	Large Language Models (LLMs) exhibit strong In-Context Learning (ICL) capabilities when prompts with demonstrations are used.
Approach:	They propose a prompt-based parameter-efficient fine-tuning approach that leverages insights into ICL’s information flow dynamics and hardwires the desired information flow into the GNN.
Outcome:	The proposed approach surpasses prompt-based fine-tuning methods in few-shot settings by updating just 0.2% to 0.5% of parameters.

Copied to clipboard

Yixin Ji, Yang Xiang, Juntao Li, Qingrong Xia, Zi Ye, Xinyu Duan, Zhefeng Wang, Kehai Chen, Min Zhang

Sparse Low-rank Adaptation of Pre-trained Language Models (2023.emnlp-main)

Challenge:	Large language models require a balance between efficiency and performance.
Approach:	They propose a low-rank compression technique that reduces non-essential parameters by decomposing weight matrices into products of two low-ranked matrici.
Outcome:	The proposed method outperforms existing pruning and low-rank compression techniques in maintaining model performance at the same compression ratio.

Copied to clipboard

Ning Ding, Xingtai Lv, Qiaosen Wang, Yulin Chen, Bowen Zhou, Zhiyuan Liu, Maosong Sun

EXIT: Context-Aware Extractive Compression for Enhancing Retrieval-Augmented Generation (2025.findings-acl)

Challenge:	Existing methods for fine-tuning pre-trained large language models in a parameter-efficient manner are gaining traction within the research community.
Approach:	They propose a method of low-rank adaptation that enables dynamic adjustments to the intrinsic rank during the adaptation process.
Outcome:	The proposed approach outperforms the current method with a fixed and unalterable intrinsic rank and a low-rank adaptation process.

Copied to clipboard

Taeho Hwang, Sukmin Cho, Soyeong Jeong, Hoyun Song, SeungYoon Han, Jong C. Park

Question Answering

Challenge:	Current retrieval-augmented generation systems struggle when retrieval models fail to rank the most relevant documents . existing extractive methods reduce latency but rely on independent, non-adaptive sentence selection .
Approach:	They introduce an extractive context compression framework that enhances retrieval-augmented generation in question answering.
Outcome:	EXIT surpasses existing compression methods and uncompressed baselines in QA accuracy . the framework reduces inference time and token count while preserving contextual dependencies .

Extract, Integrate, Compete: Towards Verification Style Reading Comprehension (2021.findings-emnlp)

Copied to clipboard

Chen Zhang, Yuxuan Lai, Yansong Feng, Dongyan Zhao

MoDification: Mixture of Depths Made Easy (2025.naacl-long)

Challenge:	VGaokao is a verification style reading comprehension dataset for Chinese language tests requiring advanced language understanding skills.
Approach:	They propose a new extract-integration-compete approach to extract complementary evidence from Chinese Language tests of Gaokao and a pairwise competition to push models to learn the subtle difference between similar text pieces.
Outcome:	The proposed approach outperforms baselines on VGaokao with retrieved complementary evidence while having the merits of efficiency and explainability.

Copied to clipboard

Chen Zhang, Meizhi Zhong, Qimeng Wang, Xuantao Lu, Zheyu Ye, Chengqiang Lu, Yan Gao, Yao Hu, Kehai Chen, Min Zhang, Dawei Song

Two Birds with One Stone: Unified Model Learning for Both Recall and Ranking in News Recommendation (2022.findings-acl)

Challenge:	Long-context efficiency is a trending topic in large language model (LLM) serving.
Approach:	They propose a method to combine long-context efficiency and mixture of depths to bring down both latency and memory.
Outcome:	The proposed method achieves 1.2 speedup in latency and 1.8 reduction in memory compared to original LLMs especially in long-context applications.

Copied to clipboard

Chuhan Wu, Fangzhao Wu, Tao Qi, Yongfeng Huang

Representation Learning

Challenge:	Existing news recommender systems conduct news recall and ranking separately with different models, but maintaining multiple models leads to high computational cost and high latency.
Approach:	They propose a unified method for recall and ranking in news recommendation that uses historical news click behaviors to extract user embeddings for ranking from the user's attention query.
Outcome:	The proposed method improves recall and ranking efficiency and effectiveness on a benchmark dataset.

SPIN: Sparsifying and Integrating Internal Neurons in Large Language Models for Text Classification (2024.findings-acl)

Copied to clipboard

Difan Jiao, Yilun Liu, Zhenwei Tang, Daniel Matter, Jürgen Pfeffer, Ashton Anderson

CORAL: Learning Consistent Representations across Multi-step Training with Lighter Speculative Drafter (2025.acl-long)

Challenge:	Large Language Models (LLMs) have revolutionized text classification, but current paradigms rely on output of final layer . implicit internal structures that contribute to LLMs' impressive performance are neglected, forgoing potential performance gains.
Approach:	They propose a model-agnostic framework that sparsifies internal neurons of intermediate layers of LLMs for text classification.
Outcome:	The proposed framework significantly improves text classification accuracy, efficiency and interpretability.

Copied to clipboard

Yepeng Weng, Dianwen Mei, Huishi Qiu, Xujie Chen, Li Liu, Jiang Tian, Zhongchao Shi

Camouflaged Chinese Spam Content Detection with Semi-supervised Generative Active Learning (2020.acl-main)

Challenge:	Existing methods that focus on training and inference suffer from misalignment . speculative decoding is a powerful technique that accelerates large language models .
Approach:	They propose a framework that improves both accuracy and efficiency in speculative drafting by using cross-step representational alignment.
Outcome:	The proposed framework outperforms existing methods on three LLM families and three benchmark datasets.

Copied to clipboard

Zhuoren Jiang, Zhe Gao, Yu Duan, Yangyang Kang, Changlong Sun, Qiong Zhang, Xiaozhong Liu

HALP: Detecting Hallucinations in Vision-Language Models without Generating a Single Token (2026.eacl-long)

Challenge:	Existing active learning models for text spam detection tasks are based on pool-based active learning, but the annotating process is laborious and time consuming for humans.
Approach:	They propose a semi-supervised active learning model to address spam imbalances . they propose masked attention learning approach and character variation graph-enhanced augmentation procedure .
Outcome:	The proposed model can improve the performance of existing models for Chinese spam detection task.

Copied to clipboard

Sai Akhil Kogilathota, Sripadha Vallabha E G, Luzhe Sun, Jiawei Zhou

Fusion meets Function: The Adaptive Selection-Generation Approach in Event Argument Extraction (2025.coling-main)

Challenge:	Existing methods for detection of hallucinations operate after text generation, making intervention costly and untimely.
Approach:	They examine whether hallucination risk can instead be predicted before any token is generated by probing a model's internal representations in a single forward pass.
Outcome:	The proposed model can detect hallucinations before token generation, while query-token representations can be more accurate.

Copied to clipboard

Guoxuan Ding, Xiaobo Guo, Xin Wang, Lei Wang, Tianshu Fu, Nan Mu, Daren Zha

From "Aha Moments" to Controllable Thinking: Toward Meta-Cognitive Reasoning in LRMs via Decoupled Reasoning and Control (2026.acl-long)

Challenge:	Event Argument Extraction is a critical subtask of Event Extraction, focused on identifying event arguments within text.
Approach:	They propose a Fusion Selection-Generation-Based Approach that merges selective and generative methods to enhance argument extraction accuracy.
Outcome:	The proposed method improves on the RAMS and WikiEvents, while preserving the unique characteristics of both methods.

Copied to clipboard

Rui Ha, Rui Pu, Chaozhuo Li, Li Sun, Sen Su

Expectation Confirmation Preference Optimization for Multi-Turn Conversational Recommendation Agent (2025.findings-acl)

Challenge:	Large Reasoning Models exhibit step-by-step reasoning, reflection, and backtracking, but these behaviors are often unregulated, leading to overthinking.
Approach:	They propose a meta-cognitive reasoning framework that decouples reasoning from control to enable independent optimization of control strategies.
Outcome:	Experiments show that the proposed model improves efficiency and accuracy across reasoning benchmarks.

Copied to clipboard

Xueyang Feng, Jingsen Zhang, Jiakai Tang, Wei Li, Guohao Cai, Xu Chen, Quanyu Dai, Yue Zhu, Zhenhua Dong

E-Verify: A Paradigm Shift to Scalable Embedding-based Factuality Verification (2025.findings-emnlp)

Challenge:	Recent advances in Large Language Models (LLMs) have propelled the development of Conversational Recommendation Agents (CRAs).
Approach:	They propose a multi-turn preference optimization paradigm that leverages Expectation Confirmation Theory to explicitly model the evolution of user satisfaction throughout multi-turned dialogues.
Outcome:	The proposed paradigm eliminates the significant sampling overhead of existing MTPO methods while ensuring the optimization process drives meaningful improvements.

Copied to clipboard

Zeyang Liu, Jingfeng Xue, Xiuqi Yang, Wenbiao Du, Jiarun Fu, Junbao Chen, Wenjie Guo, Yong Wang

CollabKG: A Learnable Human-Machine-Cooperative Information Extraction Toolkit for (Event) Knowledge Graph Construction (2024.lrec-main)

Challenge:	Existing factuality verification methods follow a Decompose-Then-Verify paradigm, which improves granularity but suffers from poor scalability and efficiency.
Approach:	They propose a Decompose-Embed-Interact paradigm that shifts factuality verification from costly text-level reasoning to efficient alignment in embedding space.
Outcome:	The proposed paradigm shifts factuality verification from costly text-level reasoning to efficient alignment in embedding space .

Copied to clipboard

Xiang Wei, Yufeng Chen, Ning Cheng, Xingyu Cui, Jinan Xu, Wenjuan Han

Quality and Efficiency of Manual Annotation: Pre-annotation Bias (2022.lrec-1)

Challenge:	Existing IE tools lack multi-task support and automatic updates for KG and EKG construction.
Approach:	They propose a human-machine-cooperative IE toolkit for KG and EKG construction that unifies different IE subtasks and integrates LLMs as the assistant machine.
Outcome:	The proposed tool improves annotation quality, efficiency, and stability simultaneously.

Copied to clipboard

Marie Mikulová, Milan Straka, Jan Štěpánek, Barbora Štěpánková, Jan Hajic

F5-TTS: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matching (2025.acl-long)

Challenge:	Annotators using pre-annotation are less efficient at producing high quality annotations.
Approach:	They propose to use an automatic pre-annotation for a task to judge annotation quality . they also evaluate the effect of automatic linguistically-based checks on the same data .
Outcome:	The proposed method improves the quality of annotated sentences without reducing quality.

Copied to clipboard

Yushen Chen, Zhikang Niu, Ziyang Ma, Keqi Deng, Chunhui Wang, JianZhao JianZhao, Kai Yu, Xie Chen

Tree-Structured Non-Autoregressive Decoding for Sequence-to-Sequence Text Generation (2025.findings-emnlp)

Challenge:	Recent research in Text-to-Speech (TTS) has experienced great advancement . current models can synthesize speech for any given text and mimic the speaker of audio prompt.
Approach:	They propose a fully non-autoregressive text-to-speech system based on flow matching with Diffusion Transformer (DiT) without complex designs such as duration model, text encoder, and phoneme alignment, the text input is simply padded with filler tokens to the same length as input speech, and then denoising is performed for speech generation.
Outcome:	The proposed system achieves an inference RTF of 0.15, which is greatly improved compared to state-of-the-art diffusion-based models.

Copied to clipboard

Pengyu Ji, Yufei Liu, Xiang Hu, Kewei Tu

Generative Deduplication For Socia Media Data Selection (2024.findings-emnlp)

Challenge:	Autoregressive Transformers suffer from high inference latency due to sequential token generation.
Approach:	They propose a tree-structured non-autoregressive decoding paradigm that bridges autoregressive and non-automatic decoding.
Outcome:	The proposed paradigm outperforms autoregressive and non-autoregressive decoding in machine translation and paraphrase generation.

Copied to clipboard

Xianming Li, Jing Li

Self-DC: When to Reason and When to Act? Self Divide-and-Conquer for Compositional Unknown Questions (2025.naacl-long)

Challenge:	Social media data exhibits severe redundancy due to its noisy nature, leading to increased training time and model bias in its processing.
Approach:	They propose a new framework for deduplication of social media data by removing semantically duplicate data from the model and add time-dimensional Gaussian noise to reduce training complexity.
Outcome:	The proposed framework can reduce training samples while improving performance over baselines.

Copied to clipboard

Hongru Wang, Boyang Xue, Baohang Zhou, Tianhua Zhang, Cunxiang Wang, Huimin Wang, Guanhua Chen, Kam-Fai Wong

Efficiently Tuned Parameters Are Task Embeddings (2022.emnlp-main)

Challenge:	Existing studies focus on leveraging internal knowledge of Large Language Models (LLMs) to answer known questions.
Approach:	They propose a framework that allows LLMs to choose between internal and external knowledge . they use a dataset to analyze compositional questions that are composed of unknown sub-questions .
Outcome:	The proposed framework can achieve comparable or even better performance with much fewer external calls compared with several strong baselines.

Copied to clipboard

Wangchunshu Zhou, Canwen Xu, Julian McAuley

ProDial – An Annotated Proactive Dialogue Act Corpus for Conversational Assistants using Crowdsourcing (2022.lrec-1)

Challenge:	Existing methods for intermediate-task transfer are computationally infeasible to experiment with all intermediate combinations.
Approach:	They propose to use task-specific parameters updated in parameter-efficient tuning methods to predict inter-task transferability.
Outcome:	The proposed approach outperforms existing methods while being conceptually simple and computationally efficient.

Copied to clipboard

Matthias Kraus, Nicolas Wagner, Wolfgang Minker

Invisible Entropy: Towards Safe and Efficient Low-Entropy LLM Watermarking (2025.emnlp-main)

Challenge:	Especially in the household domain, robots may become indispensable helpers by overtaking tedious tasks, e.g. keeping the place tidy.
Approach:	They propose a conversational approach for explicitly collecting personal user information using natural dialogue.
Outcome:	The proposed approach is compared to a baseline dialogue strategy for interactive personalization and has shown that it is friendlier.

Copied to clipboard

Tianle Gu, Zongqi Wang, Kexin Huang, Yuanqi Yao, Xiangliang Zhang, Yujiu Yang, Xiuying Chen

Tree Communication Models for Sentiment Analysis (P19-1)

Challenge:	Existing methods to watermark low-entropy content are expensive and risky . IE reduces parameter size by 99% while achieving performance on par with state-of-the-art methods .
Approach:	They propose a logit-based watermarking paradigm that uses entropy-based features to predict whether the next token is high or low.
Outcome:	The proposed method reduces parameter size by 99% while achieving performance on par with state-of-the-art methods.

Copied to clipboard

Yuan Zhang, Yue Zhang

Dynamic Feature Fusion for Sign Language Translation Using HyperNetworks (2025.findings-naacl)

Challenge:	Existing methods for sentiment classification over hierarchical phrases capture only bottom-up dependencies between constituents.
Approach:	They propose a tree-based sentiment analysis model using graph convolutional neural network and graph recurrent neural network which allows rich information exchange between phrases constituent tree.
Outcome:	The proposed model outperforms existing tree-LSTMs in accuracy and efficiency, providing more consistent predictions on phrase-level sentiments.

Copied to clipboard

Ruiquan Zhang, Rui Zhao, Zhicong Wu, Liang Zhang, Haoqi Zhang, Yidong Chen

Machine Translation

Challenge:	Using RGB and keypoint streams, sign language translation is highly dependent on the brain's ability to process color, shape, and motion simultaneously.
Approach:	They propose a hypernetwork-based fusion method that extracts salient features from RGB and keypoint streams and introduces self-distillation and SST contrastive learning to maintain feature advantages while aligning the global semantic space.
Outcome:	The proposed method achieves state-of-the-art performance on two public sign language datasets, reducing model parameters by about two-thirds.

How to Determine the Most Powerful Pre-trained Language Model without Brute Force Fine-tuning? An Empirical Survey (2023.findings-emnlp)

Copied to clipboard

Jun Bai, Xiaofeng Zhang, Chen Li, Hanhua Hong, Xi Xu, Chenghua Lin, Wenge Rong

MADD: Multi-Agent Drug Discovery Orchestra (2025.findings-emnlp)

Challenge:	Transferability estimation has been a topic of great interest in computer vision fields . a lack of a comprehensive comparison between these estimation methods is a problem .
Approach:	They conduct a thorough survey of existing methods to find the most suitable model . they also outline difficulties of consideration of training details and applicability to text generation .
Outcome:	The proposed methods perform well with superiorities in effectiveness and efficiency.

Copied to clipboard

Gleb Vitalevich Solovev, Alina Borisovna Zhidkovskaya, Anastasia Orlova, Nina Gubina, Anastasia Vepreva, Rodion Golovinskii, Ilya Tonkii, Ivan Dubrovsky, Ivan Gurev, Dmitry Gilemkhanov, Denis Chistiakov, Timur A. Aliev, Ivan Poddiakov, Galina Zubkova, Ekaterina V. Skorb, Vladimir Vinogradov, Alexander Boukhanovsky, Nikolay Nikitin, Andrei Dmitrenko, Anna Kalyuzhnaya, Andrey Savchenko

Challenge:	Recent advances in artificial intelligence have limited access to wet-lab tools for hit identification . multi-agent systems combine interpretability of LLMs with precision of specialized models and tools .
Approach:	They propose a multi-agent system that builds and executes customized hit identification pipelines from natural language queries.
Outcome:	The proposed system reduces the complexity of traditional screening methods and improves efficiency.

AutoBreach: Universal and Adaptive Jailbreaking with Efficient Wordplay-Guided Optimization via Multi-LLMs (2025.findings-naacl)

Copied to clipboard

Jiawei Chen, Xiao Yang, Zhengwei Fang, Yu Tian, Yinpeng Dong, Zhaoxia Yin, Hang Su

Personalized Transformer for Explainable Recommendation (2021.acl-long)

Challenge:	Existing jailbreak research exhibits limitations in universality, validity, and efficiency . Existing methods for jailbreaking LLMs have limited validity and effectiveness .
Approach:	They propose a black-box approach that uses wordplay-guided mapping rule sampling to create universal adversarial prompts.
Outcome:	The proposed method efficiently identifies security vulnerabilities across various LLMs, achieving an average success rate of over 80% with fewer than 10 queries.

Copied to clipboard

Lei Li, Yongfeng Zhang, Li Chen

Explainability & Interpretability in NLP

Challenge:	Recent years have witnessed the successful application of natural language generation.
Approach:	They propose a model that uses user and item IDs to predict the words in the target explanation to make personalized Transformer.
Outcome:	The proposed model outperforms BERT on the explainable recommendation task in terms of effectiveness and efficiency.

Revisiting Model Interpolation for Efficient Reasoning (2026.acl-long)

Copied to clipboard

Taiqiang Wu, Runming Yang, Tao Liu, Jiahao Wang, Ngai Wong

Compressing Context to Enhance Inference Efficiency of Large Language Models (2023.emnlp-main)

Challenge:	Existing models that interpolate weights of two specialized models can be abused for efficient reasoning.
Approach:	They propose to merge two specialized models and create a model that combines efficiency and efficiency.
Outcome:	The proposed method outperforms existing models on efficiency and effectiveness.

Copied to clipboard

Yucheng Li, Bo Dong, Frank Guerin, Chenghua Lin

Spotlighter: Revisiting Prompt Tuning from a Representative Mining View (2025.findings-emnlp)

Challenge:	Large language models (LLMs) have demonstrated remarkable power and impressive generalisation abilities across various tasks.
Approach:	They propose a method that prunes redundancies in the input context to make the input more compact.
Outcome:	The proposed method reduces memory and inference time while maintaining comparable performance compared to full context.

Copied to clipboard

Yutong Gao, Maoyuan Shao, Xinyang Huang, Chuang Zhu, Yu Weng, Xuan Liu, Lijuan Sun, Guoshun Nan

Dynamic Transformers Provide a False Sense of Efficiency (2023.acl-long)

Challenge:	Spotlighter is a lightweight token-selection framework that enhances accuracy and efficiency in prompt tuning.
Approach:	They propose a token-selection framework that enhances accuracy and efficiency in prompt tuning by preserving only the top-scoring tokens for downstream prediction.
Outcome:	The proposed framework outperforms CLIP by up to 11.19% in harmonic mean accuracy and achieves 0.8K additional FPS, with only 21 extra parameters.

Copied to clipboard

Yiming Chen, Simin Chen, Zexin Li, Wei Yang, Cong Liu, Robby Tan, Haizhou Li

Easy First Relation Extraction with Information Redundancy (D19-1)

Challenge:	Pre-trained language models typically lead to high computational cost during inference.
Approach:	They propose a slowdown attack framework that can reduce inference efficiency by 80% by leveraging existing adversarial attacks targeting model accuracy.
Outcome:	The proposed framework can reduce the efficiency of multi-exit models by 80% on average, validating its effectiveness and generalization ability.

Copied to clipboard

Shuai Ma, Gang Wang, Yansong Feng, Jinpeng Huai

Token and Head Adaptive Transformers for Efficient Natural Language Processing (2022.coling-1)

Challenge:	Existing relation extraction models make decisions globally using integer linear programming . Existing approaches require time and memory to encode redundant information for ILP .
Approach:	They propose an easy first approach for relation extraction with information redundancies embedded in local sentence extractors to resolve conflict decisions with domain and uniqueness constraints.
Outcome:	The proposed approach outperforms both ILP and neural network-based methods in relation extraction (RE) studies have shown that the proposed approach improves the efficiency and accuracy of RE models.

Copied to clipboard

Chonghan Lee, Md Fahim Faysal Khan, Rita Brugarolas Brufau, Ke Ding, Vijaykrishnan Narayanan

Granularity Matters: Pathological Graph-driven Cross-modal Alignment for Brain CT Report Generation (2023.emnlp-main)

Challenge:	Pre-trained language models like BERT have shown significant accuracy improvements on various tasks, but their computational cost and memory footprint are prohibitive.
Approach:	They propose to extend Length Adaptive Transformer to extend the model to a token and head pruning scheme to optimize pruning efficiency.
Outcome:	The proposed model can compress and accelerate BERT-based models by fine-tuning and a token and head pruning scheme.

Copied to clipboard

Yanzhao Shi, Junzhong Ji, Xiaodan Zhang, Liangqiong Qu, Ying Liu

GlotLID: Language Identification for Low-Resource Languages (2023.findings-emnlp)

Challenge:	Existing methods for automatic Brain CT reports are limited by coarse-grained supervision and coupled cross-modal alignment.
Approach:	They propose a pathological Graph-driven cross-modal alignment model that learns fine-grained visual cues and aligns them with textual words.
Outcome:	The proposed model can improve the automatic generation of Brain CT reports and contribute to improved cranial disease diagnosis.

Copied to clipboard

Amir Kargaran, Ayyoob Imani, François Yvon, Hinrich Schuetze

LVLMs and Humans Ground Differently in Referential Communication (2026.acl-long)

Challenge:	Existing web-mined datasets for low-resource languages have been useful for low resource NLP.
Approach:	They propose a model that identifies 1665 low-resource languages and a new model that is rigorously evaluated and reliable.
Outcome:	The proposed model outperforms baselines when balancing F1 and false positive rate (FPR).

Copied to clipboard

Peter Zeng, Weiling Li, Amie J. Paige, Zhengxiang Wang, Panagiotis Kaliosis, Dimitris Samaras, Gregory J. Zelinsky, Susan Brennan, Owen Rambow

Challenge:	generative AI agents cannot model common ground in a way that enables smooth communication . a recent study examined whether large language models and large vision language models engage in grounding as human discourse partners do .
Approach:	They propose to use referential communication to model common ground between a pair of directors and a picture matching system.
Outcome:	The proposed experiment shows that generative AI agents cannot model common ground . human conversation relies on common ground accrued and updated by interacting partners .

Accelerating Inference for Pretrained Language Models by Unified Multi-Perspective Early Exiting (2022.coling-1)

Copied to clipboard

Jun Kong, Jin Wang, Liang-Chih Yu, Xuejie Zhang

Budget-Aware Anytime Reasoning with LLM-Synthesized Preference Data (2026.findings-acl)

Challenge:	Existing competitive methods to accelerate inference of pretrained language models are limited by their complexity and computational consumption.
Approach:	They propose a unified horizontal and vertical multi-perspective early exiting framework to accelerate inference of transformer-based models.
Outcome:	Experiments show that MPEE can achieve higher acceleration inference with competent performance than existing competitive methods.

Copied to clipboard

Xuanming Zhang, Shwan Ashrafi, Aziza Mirsaidova, Amir H. Rezaeian, Miguel Ballesteros, Lydia Chilton, Zhou Yu, Dan Roth

OIE@OIA: an Adaptable and Efficient Open Information Extraction Framework (2022.acl-long)

Challenge:	Recent work has explored reasoning efficiency via test-time scaling and early exit strategies.
Approach:	They propose an anytime reasoning framework and the Anytime Index to improve model quality . they also propose an inference-time self-improvement method to produce better intermediate solutions .
Outcome:	The proposed method improves on NaturalPlan, AIME, and GPQA datasets and improves reasoning quality and efficiency under budget constraints.

Copied to clipboard

Xin Wang, Minlong Peng, Mingming Sun, Ping Li

AnchorAttention: Difference-Aware Sparse Attention with Stripe Granularity (2025.emnlp-main)

Challenge:	Different Open Information Extraction (OIE) tasks require different types of information.
Approach:	They propose to adapt an OIE Graph to different OIE tasks with simple rules . they implement an end-to-end OIA generator and make it open-accessible .
Outcome:	The proposed system achieves new SOTA performance on three popular OIE tasks.

Copied to clipboard

Yu Zhang, Dong Guo, Fang Wu, Guoliang Zhu, Dian Ding, Yiming Zhang

LEMON: Reviving Stronger and Smaller LMs from Larger LMs with Linear Parameter Fusion (2024.acl-long)

Challenge:	Existing methods for large language models with extended context lengths face significant computational challenges during the prefill phase.
Approach:	They propose a difference-aware, dynamic sparse attention mechanism that efficiently identifies critical attention regions at a finer stripe granularity while adapting to global contextual information.
Outcome:	The proposed model achieves a speedup of 1.44 while maintaining higher recall rates.

Copied to clipboard

Yilong Chen, Junyuan Shang, Zhenyu Zhang, Shiyao Cui, Tingwen Liu, Shuohuan Wang, Yu Sun, Hua Wu

Calibrating Factual Knowledge in Pretrained Language Models (2022.findings-emnlp)

Challenge:	Existing methods to train a stronger and smaller model with the help of large models are limited by the model size and performance.
Approach:	They propose to learn competent initial points for smaller models by fusing parameters from larger models and introduce controllable receptive fields to model prior parameter characteristics.
Outcome:	The proposed method outperforms baselines in terms of effectiveness and efficiency.

Copied to clipboard

Qingxiu Dong, Damai Dai, Yifan Song, Jingjing Xu, Zhifang Sui, Lei Li

LLM-Independent Adaptive RAG: Let the Question Speak for Itself (2025.emnlp-main)

Challenge:	Existing studies show that Pretrained Language Models can store factual knowledge, but facts stored in PLMs are not always correct.
Approach:	They propose a lightweight method to calibrate factual knowledge in PLMs without re-training from scratch.
Outcome:	The proposed method can be used to calibrate factual knowledge in PLMs without re-training from scratch.

Copied to clipboard

Maria Marina, Nikolay Ivanov, Sergey Pletenev, Mikhail Salnikov, Daria Galimzianova, Nikita Krayko, Vasily Konovalov, Alexander Panchenko, Viktor Moskvoretskii

DLPO: Towards a Robust, Efficient, and Generalizable Prompt Optimization Framework from a Deep-Learning Perspective (2025.findings-emnlp)

Challenge:	Existing methods to retrieve Large Language Models (LLMs) are inefficient and impractical.
Approach:	They propose a lightweight adaptive retrieval method that leverages external information to achieve comparable quality while achieving significant efficiency gains.
Outcome:	The proposed methods achieve comparable quality while achieving significant efficiency gains on 6 QA datasets.

Copied to clipboard

Dengyun Peng, Yuhang Zhou, Qiguang Chen, JinHao Liu, Jingjing Chen, Libo Qin, Wanxiang Che

H-STAR: LLM-driven Hybrid SQL-Text Adaptive Reasoning on Tables (2025.naacl-long)

Challenge:	Existing methods for prompt optimization still face challenges in robustness, efficiency, and generalization.
Approach:	They propose 7 new approaches inspired by traditional deep learning paradigms for prompt optimization that integrate text-based gradient optimization.
Outcome:	The proposed methods integrate deep learning paradigms into text-based gradient optimization.

Copied to clipboard

Nikhil Abhyankar, Vivek Gupta, Dan Roth, Chandan K. Reddy

SafeInt: Shielding Large Language Models from Jailbreak Attacks via Safety-Aware Representation Intervention (2025.findings-emnlp)

Challenge:	Existing methods for tabular reasoning combine textual and symbolic reasoning in a two-stage process to address these limitations.
Approach:	They propose an algorithm that integrates symbolic and semantic (textual) approaches in a two-stage process to address these limitations.
Outcome:	The proposed algorithm outperforms state-of-the-art methods across three tabular question-answering and fact-verification datasets, underscoring its effectiveness and efficiency.

Copied to clipboard

Jiaqi Wu, Chen Chen, Chunyan Hou, Xiaojie Yuan

An Empirical Study of Frame Selection for Text-to-Video Retrieval (2023.findings-emnlp)

Challenge:	Jailbreak attacks exploit vulnerabilities in large language models to induce undesirable behavior . existing defenses cannot dynamically adjust representations based on harmfulness of queries .
Approach:	They propose a representation-aware representation method that shields LLMs from jailbreak attacks . SafeInt relocates jailbreak-related representations into the rejection region .
Outcome:	The proposed method outperforms baseline defenses while maintaining utility . it relocates jailbreak-related representations into the rejection region .

Copied to clipboard

Mengxia Wu, Min Cao, Yang Bai, Ziyin Zeng, Chen Chen, Liqiang Nie, Min Zhang

AdaSwitch: Adaptive Switching between Small and Large Agents for Effective Cloud-Local Collaborative Learning (2024.emnlp-main)

Challenge:	Existing methods for text-to-video retrieval select a subset of frames to represent video content . current methods only explore video contents while ignoring relevancy to texts .
Approach:	They propose to use a subset of frames to represent video content for TVR . they analyze six different frame selection methods to determine their effectiveness .
Outcome:	The proposed method improves retrieval efficiency without sacrificing visual details . the proposed method explores the video contents while ignoring relevancy to texts .

Copied to clipboard

Hao Sun, Jiayi Wu, Hengyi Cai, Xiaochi Wei, Yue Feng, Bo Wang, Shuaiqiang Wang, Yan Zhang, Dawei Yin

Boosting Long-Context Information Seeking via Query-Guided Activation Refilling (2025.acl-long)

Challenge:	Recent advances in large language models have been remarkable . users face a choice between using cloud-based LLMs for generation quality or local-based ones for lower computational cost .
Approach:	They propose a new LLM utilization paradigm that facilitates collaborative operation . they evaluate AdaSwitch across 7 benchmarks and compare it to other LLMs .
Outcome:	The proposed model improves performance of local and cloud agents across 7 benchmarks . it achieves competitive results compared to the cloud agent while utilizing less computational overhead.

Copied to clipboard

Hongjin Qian, Zheng Liu, Peitian Zhang, Zhicheng Dou, Defu Lian

EffiQA: Efficient Question-Answering with Strategic Multi-Model Collaboration on Knowledge Graphs (2025.coling-main)

Challenge:	Existing methods for processing long contexts are ineffective due to their inherent context window limitations and the computational burden of extensive key-value activations.
Approach:	They propose a method for processing long context information-seeking tasks via query-guided ACtivation REfilling (ACRE) a bi-layer KV Cache is constructed where the layer-1 cache compactly captures global information and the layer-2 cache provides detailed, localized information.
Outcome:	The proposed method achieves significant improvements in both performance and efficiency on a variety of long-context information-seeking datasets.

Copied to clipboard

Zixuan Dong, Baoyun Peng, Yufei Wang, Jia Fu, Xiaodong Wang, Xin Zhou, Yongxue Shan, Kangchen Zhu, Weiguo Chen

Tied-LoRA: Enhancing parameter efficiency of LoRA with Weight Tying (2024.naacl-long)

Challenge:	Existing approaches that integrate LLMs and KGs either underutilize the reasoning abilities of LLM or suffer from prohibitive computational costs due to tight coupling.
Approach:	They propose a framework that can strike a balance between performance and efficiency via an iterative paradigm.
Outcome:	The proposed framework can strike a balance between performance and efficiency via an iterative paradigm.

Copied to clipboard

Adithya Renduchintala, Tugrul Konuk, Oleksii Kuchaiev

Federated Learning of Large Language Models with Parameter-Efficient Prompt Tuning and Adaptive Optimization (2023.emnlp-main)

Challenge:	a new paradigm for low-rank Adaptation (LoRA) uses weight tying and selective training to improve parameter efficiency.
Approach:	They propose a paradigm that uses weight tying and selective training to enhance parameter efficiency of Low-rank Adaptation.
Outcome:	The proposed paradigm achieves comparable performance to LoRA with reduced model complexity . the proposed paradigm can be used for a variety of tasks and languages .

Copied to clipboard

Tianshi Che, Ji Liu, Yang Zhou, Jiaxiang Ren, Jiwen Zhou, Victor Sheng, Huaiyu Dai, Dejing Dou

EEE-QA: Exploring Effective and Efficient Question-Answer Representations (2024.lrec-main)

Challenge:	Prompt tuning of Large Language Models (LLMs) can incur performance degradation or low training efficiency.
Approach:	They propose a prompt tuning approach with Adaptive Optimization to enable efficient FL of LLMs.
Outcome:	The proposed approach improves performance and efficiency simultaneously and addresses client drift problems on both the device and server sides.

Copied to clipboard

Zhanghao Hu, Yijun Yang, Junjie Xu, Yifu Qiu, Pinzhen Chen

Efficiency and Effectiveness in Task-Oriented Dialogue: On Construction Repetition, Information Rate, and Task Success (2024.lrec-main)

Challenge:	Current approaches to question answering rely on pre-trained language models like RoBERTa.
Approach:	They propose a pooling approach that embeds all answer candidates with the question . they also propose enabling cross-reference between answer choices .
Outcome:	The proposed methods improve throughput and memory efficiency with little sacrifice in performance.

Copied to clipboard

Jun Sen Yee, Mario Giulianelli, Arabella J. Sinclair

Boosting the Capabilities of Compact Models in Low-Data Contexts with Large Language Models and Retrieval-Augmented Generation (2025.coling-main)

Challenge:	Repetition of constructions in task-oriented dialogue can have negative and positive effects on information rate and delivery, but is also predictive of task success.
Approach:	They investigate the role that efficiency and effectiveness play in speakers’ repetition of shared word sequences, or constructions, in task-oriented dialogue.
Outcome:	The results show that repeating constructions has negative and positive effects on information rate and delivery and that information rate managing strategies are predictive of task success.

Copied to clipboard

Bhargav Shandilya, Alexis Palmer

Bi-Chainer: Automated Large Language Models Reasoning with Bidirectional Chaining (2024.findings-acl)

Challenge:	Existing language models lack data and computation power, but they are extremely parameter-heavy and difficult to train.
Approach:	They propose a retrieval augmented generation framework backed by a large language model to correct the output of a smaller model for morphological glossing.
Outcome:	The proposed model is highly effective in data-scarce settings and offers a state-of-the-art for morphological glossing.

Copied to clipboard

Shuqi Liu, Bowei He, Linqi Song

Tree-of-Code: A Self-Growing Tree Framework for End-to-End Code Generation and Execution in Complex Tasks (2025.findings-acl)

Challenge:	Existing unidirectional chaining methods suffer from low prediction accuracy and efficiency.
Approach:	They propose a bidirectional chaining method which dynamically switches to depth-first reasoning in the opposite reasoning direction when it encounters multiple branching options within the current direction.
Outcome:	The proposed method achieves sizable accuracy boots over unidirectional chaining frameworks on four challenging logical reasoning datasets.

Copied to clipboard

Ziyi Ni, Yifan Li, Ning Yang, Dou Shen, Pin Lyu, Daxiang Dong

CoQuIR: A Comprehensive Benchmark for Code Quality-Aware Information Retrieval (2026.acl-long)

Challenge:	Effectively and efficiently handling complex realworld problems has become a key focus across industry and academia.
Approach:	They propose a tree-of-code framework that generates nodes through self-supervision and combines prompt and model exploration in a GT-free setting.
Outcome:	Experiments on two datasets with ten popular zero-shot LLMs show that Tree-of-Code boosts accuracy by nearly 20% over CodeAct with fewer than 1/4 turns.

Copied to clipboard

Jiahui Geng, Fengyu Cai, Shaobo Cui, Qing Li, Liangwei Chen, Chenyang Lyu, Haonan Li, Derui Zhu, Alexander Pretschner, Heinz Koeppl, Fakhri Karray

LEDGER: Scaling Agentic Document Editing with Dependency-aware Graph Retrieval (2026.findings-acl)

Challenge:	Existing benchmarks focus on functional relevance while neglecting code quality.
Approach:	They propose a multilingual benchmark to evaluate quality-aware code retrieval . they include fine-grained quality annotations over 42,725 queries and 134,907 code snippets .
Outcome:	The proposed benchmarks show that state-of-the-art models fail to separate buggy or insecure code from robust counterparts.

Copied to clipboard

Hang Wang, Utkarsh Garg, Reza Davari, Huitian Jiao, Hao Cheng, Baolin Peng, Si-Qing Chen, Tao Ge

ProsodyFlow: High-fidelity Text-to-Speech through Conditional Flow Matching and Prosody Modeling with Large Speech Language Models (2025.coling-main)

Challenge:	Document editing requires full-context awareness of dependencies, but processing entire documents for each edit incurs prohibitive token costs and latency.
Approach:	a framework that constructs lightweight dependency graphs captures semantic relationships and structural hierarchies across document elements is proposed for agentic document editing . a scaLing agentic agentic framework is based on a dependency graph framework that captures dependencies and refactors function dependencies.
Outcome:	a new framework achieves 76 consistency versus 56 baseline while reducing token usage by 85 . the framework is based on a framework that captures semantic relationships and structural hierarchies across document elements . it can be used to improve document consistency, but it also reduces token costs and latency .

Copied to clipboard

Haoyu Wang, Sizhe Shan, Yinlin Guo, Yuehai Wang

Active Learning for Abstractive Text Summarization via LLM-Determined Curriculum and Certainty Gain Maximization (2024.findings-emnlp)

Challenge:	Text-to-speech (TTS) models have been developed to generate high-quality speech.
Approach:	They propose an end-to-end TTS model that integrates large self-supervised speech models and conditional flow matching to model prosodic features effectively.
Outcome:	The proposed model improves synthesis quality and efficiency compared to existing models, showing that it generates more prosodic and expressive speech synthesizing.

Copied to clipboard

Dongyuan Li, Ying Zhang, Zhen Wang, Shiyin Tan, Satoshi Kosugi, Manabu Okumura

Uncertainty-aware Parameter-Efficient Self-training for Semi-supervised Language Understanding (2023.findings-emnlp)

Challenge:	Abstractive text summarization (ATS) requires laborious data annotation and time-consuming model training.
Approach:	They propose a novel active learning framework that asks large language models to rate difficulty of instances and then uses certainty gain maximization to select instances with a distribution that aligns well with the overall distribution.
Outcome:	The proposed framework improves stability, effectiveness, and efficiency of abstractive text summarization backbones.

Copied to clipboard

Jianing Wang, Qiushi Sun, Nuo Chen, Chengyu Wang, Jun Huang, Ming Gao, Xiang Li

From Bytes to Subwords: Challenges of Input Representations in NLP (2026.findings-acl)

Challenge:	Existing methods for pre-trained language models rely on noisy data, which can be expensive if all parameters are updated.
Approach:	They propose a self-training framework that incorporates Monte Carlo dropouts into the model and judiciously selects reliable pseudo-labeled examples based on confidence and certainty.
Outcome:	The proposed framework improves performance and efficiency over multiple tasks over multiple datasets.

Copied to clipboard

Rob Van Der Goot

DetermLR: Augmenting LLM-based Logical Reasoning from Indeterminacy to Determinacy (2024.acl-long)

Challenge:	Traditionally, characters or words have been used, but recently, subwords have become the standard.
Approach:	They examine the current use of tokenizers and examine the weaknesses of character normalization . they propose proof of concept alternatives focused on fairness and efficiency .
Outcome:	The proposed model is based on a systematic review of current tokenizers and character encodings.

Copied to clipboard

Hongda Sun, Weikai Xu, Wei Liu, Jian Luan, Bin Wang, Shuo Shang, Ji-Rong Wen, Rui Yan

MoQAE: Mixed-Precision Quantization for Long-Context LLM Inference via Mixture of Quantization-Aware Experts (2025.acl-long)

Challenge:	Recent advances in large language models (LLMs) have revolutionized the landscape of reasoning tasks.
Approach:	They propose a new approach that rethinks the reasoning process as an evolution from indeterminacy to determinacy.
Outcome:	The proposed model surpasses all baselines on various logical reasoning benchmarks.

Copied to clipboard

Wei Tao, Haocheng Lu, Xiaoyang Qu, Bin Zhang, Kai Lu, Jiguang Wan, Jianzong Wang

DSG-MCTS: A Dynamic Strategy-Guided Monte Carlo Tree Search for Diversified Reasoning in Large Language Models (2025.emnlp-main)

Challenge:	Existing approaches to optimize large language models for long-context inference are inefficient and consume memory.
Approach:	They propose a mixed-precision quantization method via mixture of experts that inputs tokens into router chunk by chunk to reduce inference overhead.
Outcome:	The proposed method outperforms state-of-the-art KV cache quantization methods on multiple benchmark datasets.

Copied to clipboard

Rui Ha, Chaozhuo Li, Rui Pu, Litian Zhang, Xi Zhang, Sen Su

Mitigating Tail Narrowing in LLM Self-Improvement via Socratic-Guided Sampling (2025.naacl-long)

Challenge:	Large language models (LLMs) have shown strong potential in complex reasoning tasks, but their performance often degrades, resulting in hallucinations, errors, and logical inconsistencies.
Approach:	They propose a framework that integrates multiple reasoning strategies to expand the reasoning space and a dynamic strategy selection mechanism that adapts to the task context.
Outcome:	The proposed framework outperforms existing state-of-the-art methods on a set of reasoning benchmarks.

Copied to clipboard

Yiwen Ding, Zhiheng Xi, Wei He, Lizhuoyuan Lizhuoyuan, Yitao Zhai, Shi Xiaowei, Xunliang Cai, Tao Gui, Qi Zhang, Xuanjing Huang

Language Detoxification with Attribute-Discriminative Latent Space (2023.acl-long)

Challenge:	Large language models (LLMs) generate solutions themselves and iteratively train on filtered, high-quality rationales, but performance reaches a ceiling after a few iterations.
Approach:	They propose a strategy to improve the efficiency of sampling heavy-tailed data by using Socratic-style guidance signals to help LLMs reasoning with complex queries.
Outcome:	The proposed approach is effective on difficult queries and on held-out tasks, while requiring human supervision.

Copied to clipboard

Jin Myung Kwak, Minseon Kim, Sung Ju Hwang

AlphaOne: Reasoning Models Thinking Slow and Fast at Test Time (2025.emnlp-main)

Challenge:	Existing methods to detoxify toxic text require excessive memory, computations and time.
Approach:	They propose a method to generate toxic text using an attribute-discriminative latent space.
Outcome:	The proposed method outperforms baselines on detoxified language and dialogue generation tasks while being time- and memory-efficient.

Copied to clipboard

Junyu Zhang, Runpei Dong, Han Wang, Xuying Ning, Haoran Geng, Peihao Li, Xialin He, Yutong Bai, Jitendra Malik, Saurabh Gupta, Huan Zhang

Local Interpretation of Transformer Based on Linear Decomposition (2023.acl-long)

Challenge:	Existing monotonic scaling methods for large reasoning models are not reliable.
Approach:	They propose a universal framework for modulating reasoning progress in large reasoning models at test time.
Outcome:	The proposed framework unifies and generalizes existing monotonic scaling methods and enables flexible and dense slow-to-fast reasoning modulation.

Copied to clipboard

Sen Yang, Shujian Huang, Wei Zou, Jianbing Zhang, Xinyu Dai, Jiajun Chen

Explainability & Interpretability in NLP

Challenge:	Existing work on local explanation generation attempts to understand model dynamics on word-level or phraselevel by assigning importance scores on input features.
Approach:	They propose to interpret neural networks by linear decomposition by a Transformer model on a single input and a linear decomposing of the output to generate local explanations.
Outcome:	The proposed method achieves competitive performance in sentiment classification and machine translation, and fidelity of explanation.

Pruning Large Language Models to Intra-module Low-rank Architecture with Transitional Activations (2024.findings-acl)

Copied to clipboard

Bowen Shen, Zheng Lin, Daren Zha, Wei Liu, Jian Luan, Bin Wang, Weiping Wang

DORA: A Dual-Objective Reinforcement Learning Framework for Effective and Efficient Multimodal Agentic Search (2026.acl-long)

Challenge:	Structured pruning is a feasible solution for end-side LLM deployment . however, achieving a high compression ratio for scaled-up LLMs remains a challenge .
Approach:	They propose a task-agnostic structured pruning approach coupled with a compact Transformer architecture to prune LLMs into an intra-module low-rank architecture.
Outcome:	The proposed approach reduces transitional activations inside multi-head attention (MHA) and multi-layer perceptron (MLP) modules while preserving inter-module activations sensitive to perturbations.

Copied to clipboard

Guangming Qin, Yuhao Deng, Yukun Zhao, Zhenyang Li, Junfeng Wang, Dawei Yin, Ye Yuan, Guoren Wang, Yizhou Yan, Chengliang Chai, Lei Cao

Distilled Dual-Encoder Model for Vision-Language Understanding (2022.emnlp-main)

Challenge:	Existing methods to train large language models overlook quality of intermediate search results . existing methods often invoke search calls during reasoning, making inference inefficient .
Approach:	They propose a dual-objective reinforcement learning framework to improve search strategies of MLLMs . DORA outperforms state-of-the-art methods, achieving up to 8.4% higher accuracy .
Outcome:	The proposed model outperforms state-of-the-art methods while reducing search calls by 9.7%.

Copied to clipboard

Zekun Wang, Wenhui Wang, Haichao Zhu, Ming Liu, Bing Qin, Furu Wei

Time Course MechInterp: Analyzing the Evolution of Components and Knowledge in Large Language Models (2025.findings-acl)

Challenge:	Experimental results show that the proposed cross-modal attention distillation is crucial to the success of our framework.
Approach:	They propose a framework that distills knowledge of fusion-encoder teacher into dual-encoding student model.
Outcome:	The proposed model is competitive with the fusion-encoder teacher model in performance, but suffers from a lack of deep cross-modal interactions.

Copied to clipboard

Ahmad Dawar Hakimi, Ali Modarressi, Philipp Wicke, Hinrich Schuetze

Single-Pass, Depth-Selective Reading for Multi-Aspect Sentiment Analysis (2026.acl-long)

Challenge:	Large language models acquire and store factual knowledge for interpretability, reliability, efficiency . prior work on factual recall focused on localizing knowledge within transformer parameters .
Approach:	They analyze the evolution of factual knowledge representation in a large language model by tracking its attention heads and feed forward networks over training.
Outcome:	The proposed model acquires and stores factual knowledge over time and is adaptively trained . the proposed model can be pruned, optimized, and transparent .

Copied to clipboard

Yan Xia, Zhuangzhuang Pan, Amirrudin Kamsin, Chee Seng Chan

Sentiment Analysis

Challenge:	Existing models re-encode the sentence for each aspect or rely on static use of deep representations, leading to redundant computation and limited adaptivity.
Approach:	They propose a single-pass inference framework that encodes each sentence once to construct a reusable, depth-ordered substrate.
Outcome:	Experiments show that DABS reduces end-to-end computation by 60% in multi-aspect settings.

CUTE: A Multilingual Dataset for Enhancing Cross-Lingual Knowledge Transfer in Low-Resource Languages (2025.coling-main)

Copied to clipboard

Wenhao Zhuang, Yuan Sun

Multilinguality

Challenge:	Existing multilingual models such as XLM-R support only approximately 100-200 languages, leaving nearly 7,000 low-resource languages untapped.
Approach:	They construct and open-source a dataset of four-language corpora obtained through machine translation into Chinese, Uyghur and Tibetan.
Outcome:	The proposed dataset includes two resource-rich languages and two low-resource languages.

AskToAct: Enhancing LLMs Tool Use via Self-Correcting Clarification (2025.emnlp-main)

Copied to clipboard

Xuan Zhang, Yongliang Shen, Zhe Zheng, Linjuan Wu, Wenqi Zhang, Yuchen Yan, Qiuying Peng, Jun Wang, Weiming Lu

Comprehensive Evaluation on Lexical Normalization: Boundary-Aware Approaches for Unsegmented Languages (2025.findings-emnlp)

Challenge:	Existing tools for ambiguous and incomplete queries are limited by manual construction and lack of error correction mechanisms during multi-turn clarification.
Approach:	They propose a framework that exploits the mapping between queries and their tool invocation solutions by removing key parameters from queries while retaining them as ground truth.
Outcome:	The proposed framework outperforms existing methods while maintaining high accuracy in tool invocation.

Copied to clipboard

Shohei Higashiyama, Masao Utiyama

Challenge:	Lexical normalization research has sought to tackle the challenge of processing informal expressions in user-generated text.
Approach:	They focus on Japanese normalization and developing methods based on state-of-the-art pre-trained models .
Outcome:	The proposed methods achieve high accuracy and efficiency across multiple evaluation perspectives.

PECAN: LLM-Guided Dynamic Progress Control with Attention-Guided Hierarchical Weighted Graph for Long-Document QA (2025.findings-acl)

Copied to clipboard

Xinyu Wang, Yanzheng Xiang, Lin Gui, Yulan He

Efficient Entity Candidate Generation for Low-Resource Languages (2022.lrec-1)

Challenge:	Long-document Question Answering (QA) challenges with large-scale text and long-distance dependencies.
Approach:	They propose a method that leverages large language models to control retrieval process . they propose 'attention-based' retrieval methods that construct hierarchical graphs .
Outcome:	The proposed method achieves LLM-level performance while maintaining computational complexity comparable to RAG methods.

Copied to clipboard

Alberto Garcia-Duran, Akhil Arora, Robert West

Graph-Reward-SQL: Execution-Free Reinforcement Learning for Text-to-SQL via Graph Matching and Stepwise Reward (2025.findings-emnlp)

Challenge:	Existing approaches for cross-lingual entity linking are not suitable for English.
Approach:	They propose a candidate generation problem in cross-lingual entity linking with a focus on low-resource languages.
Outcome:	The proposed solution outperforms the state-of-the-art approach on 9 real-world datasets and query types.

Copied to clipboard

Han Weng, Puzhen Wu, Cui Longjie, Yi Zhan, Boyi Liu, Yuanfeng Song, Dun Zeng, Yingxiang Yang, Qianru Zhang, Dong Huang, Xiaoming Yin, Yang Sun, Xing Chen

Human Guided Exploitation of Interpretable Attention Patterns in Summarization and Topic Segmentation (2022.emnlp-main)

Challenge:	Existing methods to enhance performance of large language models (LLMs) on Text-to-SQL tasks rely on execution-based or LLM-based reward models.
Approach:	They propose a reward model framework for RL-based Text-to-SQL that employs the GMNScore outcome reward model.
Outcome:	The proposed reward model outperforms existing reward models on standard benchmarks including Spider and BIRD.

Copied to clipboard

Raymond Li, Wen Xiao, Linzi Xing, Lanjun Wang, Gabriel Murray, Giuseppe Carenini

MaXIFE: Multilingual and Cross-lingual Instruction Following Evaluation (2025.acl-long)

Challenge:	Existing studies have investigated the multi-head self-attention mechanism of transformers.
Approach:	They propose to use a human-in-the-loop pipeline to discover task-specific attention patterns and inject them into transformer models to improve their accuracy.
Outcome:	The proposed methods improve the performance of transformer models by incorporating predefined patterns into their attention matrices.

Copied to clipboard

Yile Liu, Ziwei Ma, Xiu Jiang, Jinglu Hu, ChangJing ChangJing, Liang Li

Multilinguality

Challenge:	Existing evaluation methods focus on single-language scenarios, overlooking multilingual and cross-lingual contexts.
Approach:	They propose a tool to assess instruction-following capabilities across 23 different languages with 1667 verifiable instruction tasks.
Outcome:	MaXIFE evaluates instruction-following capabilities across 23 languages with 1667 verifiable instruction tasks.

MedVerse: Efficient and Reliable Medical Reasoning via DAG-Structured Parallel Execution (2026.acl-long)

Copied to clipboard

Jianwen Chen, Xinyu Yang, Peng Xia, Arian Azarang, Yueh Z Lee, Gang Li, Hongtu Zhu, Yun Li, Beidi Chen, Huaxiu Yao

ClusterAttn: KV Cache Compression under Intrinsic Attention Clustering (2025.acl-long)

Challenge:	Recent advances in large reasoning models have broadened the capabilities of medical artificial intelligence.
Approach:	They propose a reasoning framework for complex medical inference that reformulates medical reasoning as a parallelizable directed acyclic graph process based on Petri Net theory.
Outcome:	The proposed reasoning framework improves strong general-purpose LLMs by up to 8.9%.

Copied to clipboard

Minwei Zhang, Haifeng Sun, Jingyu Wang, Shaolong Li, Wanyi Ning, Qi Qi, Zirui Zhuang, Jianxin Liao

CoG: Controllable Graph Reasoning via Relational Blueprints and Failure-Aware Refinement over Knowledge Graphs (2026.acl-long)

Challenge:	Existing methods for sparse attention apply the same pattern across different attention heads and inputs, but fail to capture the intrinsic attention clustering in large language models.
Approach:	They propose a training-free sparse attention method that provides an efficient prompt cache compression scheme under intrinsic attention clustering for efficient LLM inference.
Outcome:	The proposed method reduces memory usage by 10%–65% and increases throughput by 2.6–4.8 times with no accuracy loss.

Copied to clipboard

Yuanxiang Liu, Songze Li, Xiaoke Guo, Zhaoyan Gong, Qifei Zhang, Huajun Chen, Wen Zhang

Sensitivity-LoRA : Low-Load Sensitivity-Based Fine-Tuning for Large Language Models (2025.findings-emnlp)

Challenge:	Existing approaches to large language models often exhibit cognitive rigidity, causing reasoning stagnation.
Approach:	They propose a training-free framework that mimics the interplay between intuition and deliberation.
Outcome:	The proposed framework outperforms state-of-the-art approaches on three benchmarks.

Copied to clipboard

Hao Zhang, Bo Huang, Zhenjia Li, Xi Xiao, Hui Yi Leong, Zumeng Zhang, Xinwei Long, Tianyang Wang, Hao Xu

Efficient Transformer Parameter Reuse via Zero-Token Mechanism (2026.findings-acl)

Challenge:	Low-Rank Adaptation (LoRA) is a promising approach to adapting LLMs to specialized tasks . existing rank allocation techniques remain computationally inefficient and unstable .
Approach:	They propose a low-rank adapted model that approximates model weight updates using low-ranked decomposition.
Outcome:	The proposed method is limited by its uniform rank allocation to each incremental matrix . it leverages the second-order derivatives of the loss function to capture weight sensitivity .

Copied to clipboard

Guanghao Li, Wenhao Jiang, Li Shen, Ming Tang, Chun Yuan

HiCOT: Improving Neural Topic Models via Optimal Transport and Contrastive Learning (2025.findings-acl)

Challenge:	Existing approaches to scaling up parameter counts are impractical for users with limited computational resources.
Approach:	They propose a decoupled parameter cycling strategy that employs a head-tail decoupling strategy to decouple the first (head) and last (tail) layers from the parameter cycling process.
Outcome:	The proposed approach achieves superior performance under strict parameter constraints and significantly reduces computational overhead via early exits.

Copied to clipboard

Hoang Tran Vuong, Tue Le, Tu Vu, Tung Nguyen, Linh Ngo Van, Sang Dinh, Thien Huu Nguyen

Mistake Notebook Learning: Batch-Clustered Failures for Training-Free Agent Adaptation (2026.findings-acl)

Challenge:	Recent advances in neural topic models (NTMs) have improved topic quality but still face challenges: weak document-topic alignment, high inference costs due to large pretrained language models, and limited modeling of hierarchical topic structures.
Approach:	They propose a framework that integrates hierarchical clustering and contrastive learning to refine document-topic relationships using compact PLM-based embeddings.
Outcome:	The proposed framework improves topic coherence, topic performance, representation quality and computational efficiency over existing NTMs.

Copied to clipboard

Xuanbo Su, Yingfang Zhang, Hao Luo, Xiaoteng Liu, Leo Huang

Text Clustering

Challenge:	Mistake Notebook Learning (MNL) is a new memory framework for large language model agents . it allows agents to distill shared error patterns into structured "mistake notes"
Approach:	They propose a new memory framework that enables agents to self-curate generalizable guidance from batch-clustered failures.
Outcome:	The proposed framework achieves competitive performance compared to existing memory mechanisms.

Continual Gradient Low-Rank Projection Fine-Tuning for LLMs (2025.acl-long)

Copied to clipboard

Chenxu Wang, Yilin Lyu, Zicheng Sun, Liping Jing

Reorder and then Parse, Fast and Accurate Discontinuous Constituency Parsing (2022.emnlp-main)

Challenge:	Low-Rank Adaptation (LoRA) offers efficiency but constrains the model’s ability to learn new tasks and transfer knowledge due to its low-rank nature and reliance on explicit parameter constraints.
Approach:	They propose a training strategy that synergistically combines full and low-rank parameters and jointly updating within a unified low-ranked gradient subspace.
Outcome:	Extensive experiments on continual learning benchmarks show that GORP improves performance compared to state-of-the-art approaches.

Copied to clipboard

Kailai Sun, Zuchao Li, Hai Zhao

LLM-Empowered Class Imbalanced Graph Prompt Learning for Online Drug Trafficking Detection (2025.findings-acl)

Challenge:	Discontinuous constituency parsing is still being developed for its efficiency and accuracy are far behind its continuous counterparts.
Approach:	They propose to transform a discontinuous constituent tree into a pseudo-continuous one by reordering words in the sentence.
Outcome:	The proposed method can transform a discontinuous constituent tree into a pseudo-continuous one by parsing and performing actions on three classical discontinuous constituency treebanks.

Copied to clipboard

Tianyi Ma, Yiyue Qian, Zehong Wang, Zheyuan Zhang, Chuxu Zhang, Yanfang Ye

CART: A Generative Cross-Modal Retrieval Framework With Coarse-To-Fine Semantic Modeling (2025.acl-long)

Challenge:	Existing approaches to combat illicit drug trafficking are impractical due to the scarcity of labeled samples and imbalance of classes.
Approach:	They propose a Large Language Model-empowered Heterogeneous Graph Prompt Learning framework for illicit drug trafficking detection that leverages LLM to facilitate heterogeneous graph neural networks to effectively identify minority classes.
Outcome:	The proposed framework is able to identify minority classes in class-imbalanced scenarios.

Copied to clipboard

Minghui Fang, Shengpeng Ji, Jialong Zuo, Hai Huang, Yan Xia, Jieming Zhu, Xize Cheng, Xiaoda Yang, Wenrui Liu, Gang Wang, Zhenhua Dong, Zhou Zhao

CodeAgent: Enhancing Code Generation with Tool-Integrated Agent Systems for Real-World Repo-level Coding Challenges (2024.acl-long)

Challenge:	Cross-modal retrieval tasks are used to retrieve data from one modality or another based on a query from another modality.
Approach:	They propose a generative cross-modal retrieval framework based on coarse-to-fine semantic modeling . they propose combining K-Means and RQ-VAE to discretize multimodal data into token sequences that support autoregressive generation.
Outcome:	The proposed framework achieves excellent performance and efficiency in multimodal retrieval tasks.

Copied to clipboard

Kechi Zhang, Jia Li, Ge Li, Xianjie Shi, Zhi Jin

When is Tree Search Useful for LLM Planning? It Depends on the Discriminator (2024.acl-long)

Challenge:	Large Language Models excel in simple tasks such as generating standalone code units, but real-world software development often involves complex code repositories with complex dependencies and extensive documentation.
Approach:	They propose a novel LLM-based agent framework that employs external tools for effective repo-level code generation.
Outcome:	The proposed framework outperforms commercial products like Github Copilot in the humanEval benchmark and shows that it is adaptable and efficient across multiple code generation tasks.

Copied to clipboard

Ziru Chen, Michael White, Ray Mooney, Ali Payani, Yu Su, Huan Sun

LLaVE: Large Language and Vision Embedding Models with Hardness-Weighted Contrastive Learning (2025.findings-emnlp)

Challenge:	Existing methods to build language agents that can plan efficiently and accurately have not met the needs of advanced planning methods to achieve such improvements.
Approach:	They propose to use iterative correction and tree search to solve multi-step problems in a language agent framework with three components: a generator, a discriminator, and a planning method.
Outcome:	The proposed methods improve performance on two tasks, text-to-SQL parsing and mathematical reasoning, while using discriminators with 90% accuracy.

Copied to clipboard

Zhibin Lan, Liqiang Niu, Fandong Meng, Jie Zhou, Jinsong Su

MasRouter: Learning to Route LLMs for Multi-Agent Systems (2025.acl-long)

Challenge:	Existing LMM-based embedding models exhibit a high degree of overlap in similarity distribution between positive and negative pairs, making it challenging to distinguish hard negative pairs effectively.
Approach:	They propose a framework that improves the embedding model's representation learning for negative pairs based on their discriminative difficulty.
Outcome:	The proposed framework improves the embedding model's representation learning for negative pairs based on their discriminative difficulty.

Copied to clipboard

Yanwei Yue, Guibin Zhang, Boyang Liu, Guancheng Wan, Kun Wang, Dawei Cheng, Yiyan Qi

UniKER: A Unified Framework for Combining Embedding and Definite Horn Rule Reasoning for Knowledge Graph Inference (2021.emnlp-main)

Challenge:	Multi-agent systems (MAS) powered by Large Language Models (LLMs) have been demonstrated to push the boundaries of LLM capabilities, yet they often face significant costs and challenges in dynamic LLM selection.
Approach:	They propose a multi-agent system routing solution that integrates all components of MAS into a unified routing framework.
Outcome:	The proposed solution is high-performing, cost-effective, and efficient . it reduces overhead by up to 52.07 compared to current methods on HumanEval .

Copied to clipboard

Kewei Cheng, Ziqing Yang, Ming Zhang, Yizhou Sun

Efficient Document Embeddings via Self-Contrastive Bregman Divergence Learning (2023.findings-acl)

Challenge:	Knowledge graph inference has been studied extensively due to its wide applications.
Approach:	They propose a framework that restricts logical rules to be definite Horn rules and can exploit the knowledge in logical rule-based reasoning and KGE in an extremely efficient way.
Outcome:	The proposed framework can exploit the knowledge in logical rules and improve KGE in an extremely efficient way.

Copied to clipboard

Daniel Saggau, Mina Rezaei, Bernd Bischl, Ilias Chalkidis

Inflate and Shrink:Enriching and Reducing Interactions for Fast Text-Image Retrieval (2021.emnlp-main)

Challenge:	Despite recent advances in transformer-based sentence encoders, the encoding of long documents (Ks of words) is still challenging with respect to both efficiency and quality considerations.
Approach:	They propose to combine a self-contrastive siamese network and a convex neural Bregman divergence network to train longfomer-based document encoders using an unsupervised contrastive learning method.
Outcome:	The proposed model outperforms baseline models on three long document topic classification tasks from the legal and biomedical domains.

Copied to clipboard

Haoliang Liu, Tan Yu, Ping Li

TokenDrop + BucketSampler: Towards Efficient Padding-free Fine-tuning of Language Models (2023.findings-emnlp)

Challenge:	a recent study shows that late-interaction methods trade off retrieval accuracy and efficiency by exploiting cross-modal interactions only in the late stage.
Approach:	They propose an inflating and shrinking approach to exploit cross-modal interactions . they inflate code inputs and shrink code outputs to exploit interactions progressively .
Outcome:	The proposed method exploits cross-modal interactions in the late stage to achieve retrieval speed.

Copied to clipboard

Amrit Nagarajan, Anand Raghunathan

SecFormer: Fast and Accurate Privacy-Preserving Inference for Transformer Models via SMPC (2024.findings-acl)

Challenge:	Pre-training of Language Models (LMs) is a challenge due to its huge computational footprint.
Approach:	They propose a framework that improves the efficiency and accuracy of LM fine-tuning by removing padding tokens from sequences that are variable-length .
Outcome:	The proposed framework accelerates fine-tuning on diverse downstream tasks by 10.61X while producing models that are up to 1.17% more accurate compared to conventional fine-uning.

Copied to clipboard

Jinglong Luo, Yehong Zhang, Zhuo Zhang, Jiaqi Zhang, Xin Mu, Hui Wang, Yue Yu, Zenglin Xu

Accelerating Prefilling via Decoding-time Contribution Sparsity (2026.findings-acl)

Challenge:	a growing number of cloud-based inference services are relying on SMPC to protect data privacy.
Approach:	They propose a framework for Privacy-Preserving Inference for Transformer models that eliminates exponential and maximum operations in PPI without sacrificing model performance.
Outcome:	The proposed framework outperforms MPCFormer in terms of performance and efficiency . it is 3.57 and 3.58 times faster than PUMA for BERTBASE and BERTLARGE .

Copied to clipboard

Zhiyuan He, Yike Zhang, Chengruidong Zhang, Huiqiang Jiang, Yuqing Yang, Lili Qiu

Challenge:	Existing acceleration methods exploit attention score sparsity by estimating blocks with high attention scores and applying dynamic sparse attention.
Approach:	They propose a method which replaces dense attention with Triangle attention in a subset of layers to reduce the time needed to decode.
Outcome:	Experiments show that TriangleMix achieves near-lossless performance on long-context and long-constrast reasoning benchmarks while significantly improving efficiency.

Scalable Efficient Training of Large Language Models with Low-dimensional Projected Attention (2024.emnlp-main)

Copied to clipboard

Xingtai Lv, Ning Ding, Kaiyan Zhang, Ermo Hua, Ganqu Cui, Bowen Zhou

Reproducibility in NLP: What Have We Learned from the Checklist? (2023.findings-acl)

Challenge:	Existing studies have found that low-rank pre-training often compromises effectiveness.
Approach:	They propose to apply low-dimensional module only to the attention layer to improve both effectiveness and efficiency.
Outcome:	The proposed model saves 12.4% time while improving test perplexity and on downstream tasks compared with vanilla Transformer.

Copied to clipboard

Ian Magnusson, Noah A. Smith, Jesse Dodge

Token Prediction as Implicit Classification to Identify LLM-Generated Text (2023.emnlp-main)

Challenge:	Scientific progress in NLP rests on the reproducibility of researchers’ claims.
Approach:	They examine 10,405 anonymous responses to the NLP Reproducibility Checklist . they find evidence of an increase in reporting of information after the Checklist's introduction .
Outcome:	The authors find that 44% of submissions that gather new data are 5% less likely to be accepted than those that did not.

Copied to clipboard

Yutian Chen, Hao Kang, Vivian Zhai, Liangze Li, Rita Singh, Bhiksha Raj

Text Classification

Challenge:	a novel approach for identifying large language models (LLMs) involved in text generation is proposed . instead of adding an additional classification layer, we reframe the classification task as a next-token prediction task .
Approach:	They propose a novel approach for identifying large language models involved in text generation . instead of adding an additional classification layer, they reframe the task as a next-token prediction task .
Outcome:	The proposed method performs exceptionally well in the text classification task . it can distinguish distinctive writing styles among various LLMs even without an explicit classifier.

DynaThink: Fast or Slow? A Dynamic Decision-Making Framework for Large Language Models (2024.emnlp-main)

Copied to clipboard

Jiabao Pan, Yan Zhang, Chen Zhang, Zuozhu Liu, Hongwei Wang, Haizhou Li

Code over Words: Overcoming Semantic Inertia via Code-Grounded Reasoning (2026.findings-acl)

Challenge:	Large language models (LLMs) have emerged as prominent foundation models for diverse applications due to their outstanding ability to understand and generate humanlike text.
Approach:	They propose a dynamic decision-making framework that categorizes tasks into two distinct pathways: 'Fast' and 'Slow' they propose 'self-consistency' strategy to replace the straight-forward decoding method used in COT prompting .
Outcome:	The proposed method achieves more than 3% increase in accuracy with lower cost on five popular reasoning benchmarks.

Copied to clipboard

Manjie Xu, Isabella Yin, Xinyi Tu, Chi Zhang, Yixin Zhu

Local Look-Ahead Guidance via Verifier-in-the-Loop for Automated Theorem Proving (2025.findings-acl)

Challenge:	Large Language Models struggle with semantic inertia, a problem that is often attributed to natural language encoding, which entangles descriptive semantics and logical rules, leading to persistent hallucinations of familiar physics despite explicit contradictory rules.
Approach:	They propose a framework that decouples logical dynamics from visual priors via amortized theory induction and counterfactual contrastive alignment.
Outcome:	The proposed framework outperforms expensive inference-time search methods in both efficiency and accuracy.

Copied to clipboard

Sara Rajaee, Kumar Pratik, Gabriele Cesa, Arash Behboodi

Adapting Where It Matters: Depth-Aware Adaptation for Efficient Multilingual Speech Recognition in Low-Resource Languages (2026.findings-acl)

Challenge:	Recent methods for AI reasoning require applying variants of reinforcement learning (RL) on rolled out trajectories, even for step-wise rewards, or large quantities of human-annotated trajectory data.
Approach:	They propose a verifier-in-the-loop design that uses an automated verifier to give intermediate feedback at each step of the reasoning process.
Outcome:	The proposed model improves on the Automatic Theorem Proving task using Lean as the verifier.

Copied to clipboard

Yang Xiao, Eun-Jung Holden, Ting Dang

Finetuning Pretrained Transformers into RNNs (2021.emnlp-main)

Challenge:	Recent speech foundation models excel at multilingual automatic speech recognition (ASR) for high-resource languages, but their performance drops substantially on low-resourced languages due to the limited data availability.
Approach:	They propose a Depth-Aware Model Adaptation framework that allocates adaptation capacity according to each layer’s role.
Outcome:	The proposed framework matches or surpasses state-of-the-art accuracy with 80% fewer trainable parameters and achieves 29% error reduction under extreme data scarcity.

Copied to clipboard

Jungo Kasai, Hao Peng, Yizhe Zhang, Dani Yogatama, Gabriel Ilharco, Nikolaos Pappas, Yi Mao, Weizhu Chen, Noah A. Smith

Large Language Models Are Better Adversaries: Exploring Generative Clean-Label Backdoor Attacks Against Text Classifiers (2023.findings-emnlp)

Challenge:	Efficient transformers outperform recurrent neural networks in natural language generation, but this comes with significant computational cost and memory footprint during generation.
Approach:	They propose to convert a pretrained transformer into its efficient recurrent counterpart, improving efficiency while maintaining accuracy.
Outcome:	The proposed transformers outperform recurrent neural networks in natural language generation but come with significant computational and memory footprint during generation.

Copied to clipboard

Wencong You, Zayd Hammoudeh, Daniel Lowd

Robustness in NLP

Challenge:	Backdoor attacks manipulate model predictions by inserting malicious "poison" instances that contain a specific pattern or "trigger."
Approach:	They propose an attack that inserts style-based triggers into training and test data by using a poison selection technique to improve the effectiveness of both LLMBkd and existing backdoor attacks.
Outcome:	The proposed attack achieves high success rates across a wide range of styles with little effort and no model training.

TARE: Lightweight Token-Aware Representation Editing for Fine-tuning Transformer-like Models (2026.acl-long)

Copied to clipboard

Yulong Wang, Siyu Zhao

LLMaAA: Making Large Language Models as Active Annotators (2023.findings-emnlp)

Challenge:	Existing PEFT methods can be costly and underfit token-level contexts.
Approach:	They propose a PEFT method that performs fine-grained, token-specific edits with a small additional inference overhead and minimal tuning.
Outcome:	The proposed method outperforms state-of-the-art methods in 8 tasks and GLUE with a minimal tuning overhead and inference overhead.

Copied to clipboard

Ruoyu Zhang, Yanzeng Li, Yongliang Ma, Ming Zhou, Lei Zou

Low-Rank Interconnected Adaptation across Layers (2025.findings-acl)

Challenge:	Existing supervised learning methods in natural language processing require large amounts of data.
Approach:	They propose an active learning loop that takes LLMs as annotators and puts them into an active loop to determine what to annotate efficiently.
Outcome:	The proposed model outperforms existing models with few-shot performance in two NLP tasks.

Copied to clipboard

Yibo Zhong, Jinman Zhao, Yao Zhou

Cut Your Losses! Learning to Prune Paths Early for Efficient Parallel Reasoning (2026.acl-long)

Challenge:	Low-rank adaptation (LoRA) is a parameter-efficient fine-tuning (PEFT) method that learns weight updates W = AB for pretrained weights W through low-rank adapters A and B.
Approach:	They propose a low-rank interconnected adaptation across layers method that introduces an interconnected framework with locally shared A and globally shared B experts.
Outcome:	The proposed method improves expressiveness across domains and modalities and enables higher-rank W with equal or fewer parameters.

Copied to clipboard

Jiaxi Bi, Tongxu Luo, Wenyu Du, Zhengyang Tang, Benyou Wang

Early Exit with Disentangled Representation and Equiangular Tight Frame (2023.findings-acl)

Challenge:	Parallel reasoning enhances Large Reasoning Models but incurs prohibitive costs due to futile paths caused by early errors.
Approach:	They propose a systematic taxonomy of path pruning to categorize methods by signal source and learnability.
Outcome:	The proposed model improves LRMs but incurs prohibitive costs due to futile paths caused by early errors.

Copied to clipboard

Yixin Ji, Jikai Wang, Juntao Li, Qiang Chen, Wenliang Chen, Min Zhang

Representation Learning

Challenge:	Existing early exit paradigm relies on training parametrical internal classifiers to complete specific tasks.
Approach:	They propose a method to decouple two distinct types of representation and introduce a non-parametric tight frame classifier for improvement.
Outcome:	Experiments on monolingual and multilingual tasks show that the proposed method improves over existing methods.

Multi-Docker-Eval: A ‘Shovel of the Gold Rush’ Benchmark on Automatic Environment Building for Software Engineering (2026.findings-acl)

Copied to clipboard

Kelin Fu, Tianyu Liu, Zeyu Shang, Yingwei MA, Jiaheng Liu, Jian Yang, Kaigui Bian

Nested-Refinement Metamorphosis: Reflective Evolution for Efficient Optimization of Networking Problems (2025.findings-acl)

Challenge:	Automated environment configuration is a critical bottleneck in scaling software engineering (SWE) automation.
Approach:	They propose a reliable evaluation standard for automated environment configuration for 40 real-world repositories spanning 9 programming languages.
Outcome:	The proposed benchmark includes 40 real-world repositories spanning 9 programming languages and measures success in achieving executable states and efficiency under realistic constraints.

Copied to clipboard

Shuhan Guo, Nan Yin, James Kwok, Quanming Yao

HiGoE: Hierarchical Graph of Evidence to Enhance Retrieval-Augmented Generation for Long-context Summarization (2026.acl-long)

Challenge:	Large Language Models (LLMs) excel in network algorithm design but suffer from inefficient iterative coding and high computational costs.
Approach:	They propose a method to iteratively refine task descriptions and metamorphosis on algorithms to generate more effective solutions.
Outcome:	Experimental results show that Nested-Refinement Metamorphosis outperforms state-of-the-art approaches in performance and efficiency.

Copied to clipboard

Long Yuan, Kaiwen Tian, Zi Chen, Bolong Zheng, Chuan Ma

Text Generation

Challenge:	Existing methods for long-context summarization fail to capture high-level thematic structures and long-range dependencies.
Approach:	They propose a hierarchical Graph of Evidence to reduce hallucination and attention dilution by replacing unreliable chunk-based methods with a filtered proposition–evidence graph.
Outcome:	Experiments show that HiGoE surpasses baselines in quality and efficiency.

CMR Scaling Law: Predicting Critical Mixture Ratios for Continual Pre-training of Language Models (2024.emnlp-main)

Copied to clipboard

Jiawei Gu, Zacc Yang, Chuanghao Ding, Rui Zhao, Fei Tan

Exploring Space Efficiency in a Tree-based Linear Model for Extreme Multi-label Classification (2024.emnlp-main)

Challenge:	Large Language Models (LLMs) excel in diverse tasks but often underperform in specialized fields due to limited domain-specific or proprietary corpus.
Approach:	They propose a power-law relationship between loss, mixture ratio, and training tokens scale and formalize the trade-off between general and domain-specific capabilities.
Outcome:	The proposed model achieves the desired domain transfer while maintaining general ability and highest utilization of available resources.

Copied to clipboard

He-Zhe Lin, Cheng-Hung Liu, Chih-Jen Lin

Faster In-Context Learning for LLMs via N-Gram Trie Speculative Decoding (2025.emnlp-main)

Challenge:	Extreme multi-label classification (XMC) aims to identify relevant subsets from numerous labels.
Approach:	They propose to store a tree model under the assumption of sparse data under the condition that some features may be unused when training binary classifiers in a trees method.
Outcome:	The proposed method can save 10% of the size of the standard one-vs-rest method for multi-label classification.

Copied to clipboard

Jinglin Chen, Qiwei Li, Zuchao Li, Baoyuan Qi, Liu Guoming, Haojun Ai, Hai Zhao, Ping Wang

LLM Agents for Coordinating Multi-User Information Gathering (2025.findings-acl)

Challenge:	In-Context Learning (ICL) is a key method in prompt engineering, but its long retrieved contexts and limited token throughput will slow reasoning speeds.
Approach:	They propose a method that leverages the overlap between context and model output to generate drafts from the context.
Outcome:	The proposed method achieves the highest mean speedup on Vicuna-7B, Llama2-7B-Chat, and Llma3-8B-Instruct tasks.

Copied to clipboard

Harsh Jhamtani, Jacob Andreas, Benjamin Van Durme

Less is More: Improving LLM Reasoning with Minimal Test-Time Intervention (2026.acl-long)

Challenge:	Recent large language models (LLMs) are becoming a crucial building block in developing automated agents that can assist human users with complex tasks.
Approach:	They introduce PeopleJoin, a benchmark for evaluating LM-mediated collaborative problem solving.
Outcome:	The proposed benchmarks are adapted from existing benchmarks for database question answering and multi-document summarization.

Copied to clipboard

Zhen Yang, Mingyang Zhang, Feng Chen, Ganggui Ding, Liang Hou, Xin Tao, Ying-Cong Chen

Zero-shot Graph Reasoning via Retrieval Augmented Framework with LLMs (2025.findings-emnlp)

Challenge:	Recent advances in large language models (LLMs) have focused on test-time scaling to improve reasoning quality but at the cost of efficiency.
Approach:	They propose a training-free framework that enhances reasoning accuracy and stability with minimal overhead.
Outcome:	The proposed framework yields consistent gains across general, coding, and STEM tasks while remaining highly efficient.

Copied to clipboard

Hanqing Li, Sharika Mahadevan, Kiran Jyothi Sheena, Henry Liang, Diego Klabjan

Towards A Unified View of Sparse Feed-Forward Network in Pretraining Large Language Model (2023.emnlp-main)

Challenge:	Existing methods that require extensive finetuning or depend on predefined algorithms are limited by training.
Approach:	a new retrieval-augmented framework is proposed that harnesses retrieval and large language models to address graph reasoning tasks.
Outcome:	The proposed method achieves 100% accuracy on most graph reasoning tasks while maintaining consistent token costs regardless of graph sizes.

Copied to clipboard

Zeyu Liu, Tim Dettmers, Xi Lin, Veselin Stoyanov, Xian Li

FacLens: Transferable Probe for Foreseeing Non-Factuality in Fact-Seeking Question Answering of Large Language Models (2025.emnlp-main)

Challenge:	Large and sparse feed-forward layers (S-FFN) have proven effective in scaling up the model size for pretraining large language models.
Approach:	They compare S-FFN architectures for language modeling and compare their performance and efficiency . they found a simpler selection method that selects blocks through their mean aggregated hidden states .
Outcome:	The proposed model size and selection method achieve lower perplexity in language model pretraining compared to existing MoE architectures.

Copied to clipboard

Yanling Wang, Haoyang Li, Hao Zou, Jing Zhang, Xinlei He, Qi Li, Ke Xu

Group-SAE: Efficient Training of Sparse Autoencoders for Large Language Models via Layer Groups (2025.emnlp-main)

Challenge:	Existing non-factuality detection methods require response generation, which incurs significant computational overhead.
Approach:	They propose a lightweight model called Factuality Lens which effectively probes hidden representations of fact-seeking questions for the NFP task.
Outcome:	The proposed model is able to probe hidden representations of fact-seeking questions and reduce development costs.

Copied to clipboard

Davide Ghilardi, Federico Belotti, Marco Molinari, Tao Ma, Matteo Palmonari

Defending LLMs against Jailbreaking Attacks via Backtranslation (2024.findings-acl)

Challenge:	Sparse Autoencoders (SAEs) are a promising unsupervised approach for understanding the representations of layers of Large Language Models (LLMs).
Approach:	They propose a method that groups similar models and trains a single SAE per group based on representational similarity across layers.
Outcome:	Experiments on Pythia family models show that the proposed method significantly accelerates training with minimal impact on reconstruction quality and comparable downstream task performance and interpretability over baseline SAEs trained layer by layer.

Copied to clipboard

Yihan Wang, Zhouxing Shi, Andrew Bai, Cho-Jui Hsieh

Reversal of Thought: Enhancing Large Language Models with Preference-Guided Reverse Reasoning Warm-up (2025.acl-long)

Challenge:	Recent advancement in large language models (LLMs) has shown their extensive applications and transformative potential to reshape people's lives.
Approach:	They propose a method which uses backtranslation to infer an input prompt from an input input prompt and then run it again on the backtranslated prompt.
Outcome:	The proposed method outperforms baselines and has little impact on the generation quality for benign input prompts.

Copied to clipboard

Jiahao Yuan, Dehui Du, Hao Zhang, Zixiang Di, Usman Naseem

PlanGenLLMs: A Modern Survey of LLM Planning Capabilities (2025.acl-long)

Challenge:	Existing methods to improve LLMs’ logical capabilities involve traceable or verifiable logical sequences that generate more reliable responses yet increase computational costs, or introduce rigid logic template rules, reducing flexibility.
Approach:	They propose a plug-and-play reasoning framework that enhances LLMs' logical reasoning abilities during the warm-up phase prior to batch inference.
Outcome:	The proposed framework surpasses baselines in both reasoning accuracy and efficiency.

Copied to clipboard

Hui Wei, Zihao Zhang, Shenghua He, Tian Xia, Shijia Pan, Fei Liu

Challenge:	Existing studies have focused on developing LLMs to automate complex planning tasks.
Approach:	They propose to provide a comprehensive overview of current LLM planners to fill this gap . they examine performance criteria including completeness, executability, optimality, representation, generalization, and efficiency .
Outcome:	The proposed survey examines performance criteria for LLM planners and highlights their strengths and weaknesses.

ACING: Actor-Critic for Instruction Learning in Black-Box LLMs (2025.emnlp-main)

Copied to clipboard

Salma Kharrat, Fares Fourati, Marco Canini

EHR-SeqSQL : A Sequential Text-to-SQL Dataset For Interactively Exploring Electronic Health Records (2024.findings-acl)

Challenge:	Large Language Models (LLMs) have impressive capabilities across tasks like classification, summarization, and reasoning.
Approach:	They propose an actor-critic reinforcement learning framework that formulates instruction optimization as a stateless, continuous-action problem.
Outcome:	The proposed framework outperforms human-written prompts in 76% of instruction-induction tasks with gains of 33 points and 10-point improvement over baseline.

Copied to clipboard

Jaehee Ryu, Seonhee Cho, Gyubok Lee, Edward Choi

Searching for Best Practices in Retrieval-Augmented Generation (2024.emnlp-main)

Challenge:	EHR-SeqSQL is the first text-to-SQl dataset to include sequential and contextual questions.
Approach:	They propose a sequential text-to-SQL dataset for electronic health records databases that addresses critical yet underexplored aspects in text- to-SqL parsing.
Outcome:	The proposed dataset improves compositional generalization efficiency and improves interactivity and compositionality.

Copied to clipboard

Xiaohua Wang, Zhenghua Wang, Xuan Gao, Feiran Zhang, Yixin Wu, Zhibo Xu, Tianyuan Shi, Zhengyuan Wang, Shizheng Li, Qi Qian, Ruicheng Yin, Changze Lv, Xiaoqing Zheng, Xuanjing Huang

Beyond Prompting: An Efficient Embedding Framework for Open-Domain Question Answering (2025.acl-long)

Challenge:	Retrieval-augmented generation (RAG) techniques have proven to be effective in integrating up-to-date information, mitigating hallucinations, and enhancing response quality, especially in specialized domains.
Approach:	They propose several strategies for deploying RAG that balance performance and efficiency.
Outcome:	The proposed approaches can significantly enhance question-answering capabilities and accelerate the generation of multimodal content using a “retrieval as generation” strategy.

Copied to clipboard

Zhanghao Hu, Hanqi Yan, Qinglin Zhu, Zhenyi Shen, Yulan He, Lin Gui

WebClipper: Efficient Evolution of Web Agents with Graph-based Trajectory Pruning (2026.acl-long)

Challenge:	Large language models (LLMs) have recently pushed open-domain question answering (ODQA) to new heights.
Approach:	They propose an embedding-level framework that enhances both the retriever and the reader by reordering query representations via lightweight linear layers under an unsupervised contrastive learning objective.
Outcome:	The proposed framework outperforms baselines in accuracy and efficiency across three open-source LLMs, three retrieval methods, and four ODQA benchmarks.

Copied to clipboard

Junjie Wang, Zequn Xie, Dan Yang, Jie Feng, Yue Shen, Duolin Sun, Meixiu Long, Yihan Jiao, Zhehao Tan, Jian Wang, Peng Wei, Jinjie Gu

MemBench: Towards More Comprehensive Evaluation on the Memory of LLM-based Agents (2025.findings-acl)

Challenge:	Open-source web agents rely on long tool-call trajectories with cyclic reasoning loops and exploration of unproductive branches.
Approach:	They propose a framework that compresses web agent trajectories via graph-based pruning.
Outcome:	The proposed framework reduces tool-call rounds by 20% while improving accuracy and efficiency while maintaining the same level of performance as existing models.

Copied to clipboard

Haoran Tan, Zeyu Zhang, Chen Ma, Xu Chen, Quanyu Dai, Zhenhua Dong

Challenge:	Recent studies have highlighted the significance of memory mechanisms in LLM-based agents, which enable them to store observed information and adapt to dynamic environments.
Approach:	They propose a dataset and benchmark to evaluate the memory capability of LLM-based agents from multiple aspects including their effectiveness, efficiency, and capacity.
Outcome:	The proposed benchmark incorporates factual memory and reflective memory as different levels, and proposes participation and observation as various interactive scenarios.

LongAttn: Selecting Long-context Training Data via Token-level Attention (2025.findings-acl)

Copied to clipboard

Longyun Wu, Dawei Zhu, Guangxiang Zhao, Zhuocheng Yu, Junfeng Ran, Xiangyu Wong, Lin Sun, Sujian Li

Probing Semantic Routing in Large Mixture-of-Expert Models (2025.findings-emnlp)

Challenge:	Existing methods to select long-context data often rely on sentence-level analysis, which can be greatly optimized in both performance and efficiency.
Approach:	They propose a token-level framework which quantifies long-range dependencies for LLMs by calculating token-based dependency strength and distribution uniformity of token scores.
Outcome:	The proposed framework quantifies long-range dependencies, enabling more accurate and efficient data selection.

Copied to clipboard

Matthew Lyle Olson, Neale Ratzlaff, Musashi Hinck, Man Luo, Sungduk Yu, Chendi Xue, Vasudev Lal

Towards Interpretable and Efficient Automatic Reference-Based Summarization Evaluation (2023.emnlp-main)

Challenge:	large mixture-of-expert models have become increasingly common in the open domain . prior work has explored functional differentiation through routing behavior .
Approach:	They investigate whether expert routing in large mixture-of-expert models is influenced by the semantics of the inputs.
Outcome:	The results show that expert routing is influenced by the semantics of the inputs.

Copied to clipboard

Yixin Liu, Alexander Fabbri, Yilun Zhao, Pengfei Liu, Shafiq Joty, Chien-Sheng Wu, Caiming Xiong, Dragomir Radev

FLARE: Fine-Grained Length-Aware Routing for Resource-Efficient Heterogeneous LLM Serving (2026.acl-long)

Challenge:	Compared to neural systems, automatic metrics should be interpretable and provide intuitive insights into system performance and output quality.
Approach:	They propose to use a two-stage evaluation pipeline to extract basic information units from one text sequence and check the extracted units in another sequence.
Outcome:	The proposed metrics can provide high interpretability at both the fine-grained unit level and summary level, and one-stage metrics that achieve a balance between efficiency and interpretability.

Copied to clipboard

Yujia Fu, Heming Zhong, Dan Huang, Yutong Lu

Targeting the Needle, Ignoring the Haystack: Anchoring Crucial Cues for Evolving Scam Call Detection via an LLM-Assisted Classifier (2026.findings-acl)

Challenge:	Existing routers lack fine-grained resource awareness across deployment settings, which degrades efficiency metrics in real-world serving.
Approach:	They propose a length-centric, resource-aware multi-LLM routing framework that uses length-based models to estimate per-query latency and cost.
Outcome:	Experiments show that FLARE reduces latency and cost by up to 68% and 75% while maintaining competitive accuracy.

Copied to clipboard

Tong Wu, Qinliang Su, Jianxing Yu, Bo Liang, Minhua Huang

Composable Text Controls in Latent Space with ODEs (2023.emnlp-main)

Challenge:	Existing methods for fraud detection on online service platforms often fail to generalize due to the scarcity of labeled data and the continuous evolution of conversational contexts.
Approach:	They propose a framework that anchors detection on Semantic Primitives . they prioritize stable evidence over conversational noise to ensure a verifiable fraud tactic .
Outcome:	The proposed framework achieves superior robustness and efficiency compared to baselines . it prioritizes stable evidence over diverse conversational noise .

Copied to clipboard

Guangyi Liu, Zeyu Feng, Yuan Gao, Zichao Yang, Xiaodan Liang, Junwei Bao, Xiaodong He, Shuguang Cui, Zhen Li, Zhiting Hu

Red Teaming Large Reasoning Models (2026.acl-long)

Challenge:	Existing approaches to composable text operations often require plug-and-play . a single LM can perform arbitrary text operation composition in the latent space .
Approach:	They propose an efficient approach for composable text operations in the latent space of text . they connect pretrained LMs to the laten space and adapt them to the space .
Outcome:	The proposed approach improves on existing methods in the latent space of text.

Copied to clipboard

Jiawei Chen, Yang Yang, Chao Yu, Yu Tian, Zhi Cao, Xue Yang, Linghao Li, Hang Su, Zhaoxia Yin

World Modeling Makes a Better Planner: Dual Preference Optimization for Embodied Task Planning (2025.acl-long)

Challenge:	Large Reasoning Models (LRMs) have emerged as a powerful advancement in multi-step reasoning tasks, but they introduce safety and reliability risks, such as CoT-hijacking and prompt-induced inefficiencies.
Approach:	They propose a unified benchmark to assess the trustworthiness of Large Reasoning Models.
Outcome:	The proposed benchmark evaluates truthfulness, safety and efficiency on 26 models.

Copied to clipboard

Siyin Wang, Zhaoye Fei, Qinyuan Cheng, Shiduo Zhang, Panpan Cai, Jinlan Fu, Xipeng Qiu

myMediCon: End-to-End Burmese Automatic Speech Recognition for Medical Conversations (2024.lrec-main)

Challenge:	Existing approaches focus on action selection or use pre-trained models as world models to enhance planning capabilities.
Approach:	They propose a new learning framework that optimizes state prediction and action selection through preference learning.
Outcome:	The proposed method outperforms existing methods and GPT-4o on VoTa-Bench and Qwen2-VL (7B), LLaVA-1.6 (7B) and LLama-3.2 (11B).

Copied to clipboard

Hay Man Htun, Ye Kyaw Thu, Hutchatai Chanlekha, Kotaro Funakoshi, Thepchai Supnithi

Speech & Audio in NLP

Challenge:	Existing medical conversation speech corpora for Burmese are limited, despite advances in ASR.
Approach:	They propose to use a manually curated medical conversation speech corpus for Burmese to examine the performance of ASR models.
Outcome:	The proposed model outperforms the Transformer model and the Recurrent Neural Network (RNN) models.

Automated CAD Modeling Sequence Generation from Text Descriptions via Transformer-Based Large Language Models (2025.acl-long)

Copied to clipboard

JianXing Liao, Junyan Xu, Yatao Sun, Maowen Tang, Sicheng He, Jingxian Liao, Shui Yu, Yun Li, Xiaohong Guan

Automated Creativity Evaluation of Language Models Across Open-Ended Tasks (2026.acl-long)

Challenge:	Experimental results demonstrate that the proposed approach outperforms traditional methods in both accuracy and efficiency.
Approach:	They propose a language-guided framework that integrates large language models with computer-automated design to address these challenges.
Outcome:	The proposed framework outperforms traditional methods in accuracy and efficiency, providing a powerful tool for automating industrial workflows and generating complex CAD models from textual prompts.

Copied to clipboard

Tan Min Sen, Zachary Choy Kit Chun, Syed Ali Redha Alsagoff, Nadya Yuki Wangsajaya, Banerjee Mohor, Swaagat Bikash Saikia, Alvin Chan

Gradient-Guided Multi-Judge Prompt Optimization (2026.acl-long)

Challenge:	Existing methods for evaluating creativity are tightly coupled to specific tasks and limiting scalability and generality.
Approach:	They propose a domain-agnostic framework for quantifying LLM creativity across open-ended tasks.
Outcome:	The proposed framework captures key facets of creativity including novelty, diversity, and task fulfilment with over 60% improved efficiency.

Copied to clipboard

ChenZhuo Zhao, Xinda Wang, Pu Zhao, Yue Huang, Junting Lu, Ziqian Liu, Qingwei Lin, Saravan Rajmohan, Dongmei Zhang

Mitigating Lost-in-Retrieval Problems in Retrieval Augmented Multi-Hop Question Answering (2025.acl-long)

Challenge:	Existing approaches to prompt optimization trade off signal quality against computational cost.
Approach:	They propose a framework that uses a first-order gradient approximation to score segment importance in a continuous masking direction.
Outcome:	The proposed framework improves efficiency and robustness by using a first-order gradient approximation to score segment importance in a continuous masking direction.

Copied to clipboard

Rongzhi Zhu, Xiangyu Liu, Zequn Sun, Yiwei Wang, Wei Hu

Text-Guided Multi-Scale Frequency Representation Adaptation (2026.acl-long)

Challenge:	Empirical results show that ChainRAG consistently outperforms baselines in both effectiveness and efficiency.
Approach:	They propose a method which sequentially handles each sub-question by completing missing key entities and retrieving relevant sentences from a sentence graph for answer generation.
Outcome:	The proposed method outperforms baselines on three multi-hop QA datasets.

Copied to clipboard

Weicai Yan, Xinhua Ma, Wang Lin, Tao Jin

Think Less, Know More: State-Aware Reasoning Compression with Knowledge Guidance for Efficient Reasoning (2026.findings-acl)

Challenge:	Existing methods for fine-tuning visual signals are limited by their size and complexity.
Approach:	They propose a multi-scale frequency-based fine-tuning method that integrates textual information and performs multi-level fine- tuning of visual signals in the frequency domain.
Outcome:	Extensive experiments on multimodal models, including CLIP and LLaVA, demonstrate that the proposed method significantly improves performance and efficiency with minimal cost and fast convergence within one epoch.

Copied to clipboard

Yi Sui, Chaozhuo Li, Dawei Song

AgentGL: Towards Agentic Graph Learning with LLMs via Reinforcement Learning (2026.acl-long)

Challenge:	Existing CoT compression methods struggle to balance accuracy and efficiency . long CoT reasoning also introduces an overthinking phenomenon, authors say .
Approach:	They propose a framework that performs step-wise CoT compression by modeling stage-specific redundancy sources and integrating with a retrieval-augmented guidance.
Outcome:	The proposed framework reduces average response length by 59.9% while improving accuracy by 4.8 points over existing methods.

Copied to clipboard

Yuanfu Sun, Kang Li, Dongzhe Fan, Jiajin Liu, Qiaoyu Tan

LongLLaVA: Scaling Multi-modal LLMs to 1000 Images Efficiently via a Hybrid Architecture (2025.findings-emnlp)

Challenge:	Existing agentic frameworks treat external information as unstructured text and fail to leverage topological dependencies inherent in real-world data.
Approach:	They propose to reframe graph learning as an interleaved process of topology-aware navigation and LLM-based inference.
Outcome:	The proposed framework outperforms strong GraphLLMs and GraphRAG benchmarks in multiple LLM backbones.

Copied to clipboard

Xidong Wang, Dingjie Song, Shunian Chen, Junying Chen, Zhenyang Cai, Chen Zhang, Lichao Sun, Benyou Wang

Yang Zhao, Li Du, Xiao Ding, Yangou Ouyang, Hepeng Wang, Kai Xiong, Jinglong Gao, Zhouhao Sun, Dongliang Xu, Qing Yang, Dongchen Li, Bing Qin, Ting Liu

Challenge:	Long-context Large Language Models (MLLMs) are critical for video understanding and image analysis.
Approach:	They propose a hybrid architecture that integrates Mamba and Transformer blocks . they introduce data construction methods that capture both temporal and spatial dependencies .
Outcome:	The proposed model achieves competitive results across various benchmarks while maintaining high throughput and low memory consumption.

Copied to clipboard

Rethinking Text-based Protein Understanding: Retrieval or LLM? (2025.emnlp-main)

Challenge:	Existing methods for selecting training data from general datasets fail to account for the joint distribution of instructions, resulting in inefficient learning and suboptimal knowledge transfer.
Approach:	They propose a method that constructs a mixed gradient-based instruction graph to capture the joint distribution and interdependencies among instructions.
Outcome:	The proposed method outperforms existing methods on domain adaptation tasks and in complex, data-scarce scenarios.

Copied to clipboard

Juntong Wu, Zijing Liu, He Cao, Li Hao, Bin Feng, Zishan Shu, Ke Yu, Li Yuan, Yu Li

The LLM Effect: Are Humans Truly Using LLMs, or Are They Being Influenced By Them Instead? (2024.emnlp-main)

Challenge:	Recent studies have focused on integrating protein-related knowledge into large language models through continued pretraining and multi-modal alignment.
Approach:	They propose a retrieval-enhanced method which significantly outperforms fine-tuned LLMs for protein-to-text generation and shows accuracy and efficiency in training-free scenarios.
Outcome:	The proposed method significantly outperforms fine-tuned LLMs for protein-to-text generation and shows accuracy and efficiency in training-free scenarios.

Copied to clipboard

Alexander Choi, Syeda Sabrina Akter, J.p. Singh, Antonios Anastasopoulos

LADR: Locality-Aware Dynamic Rescue for Efficient Text-to-Image Generation with Diffusion Large Language Models (2026.acl-long)

Challenge:	Large language models have shown capabilities close to human performance in various analytical tasks.
Approach:	They investigate the efficiency and accuracy of Large Language Models in specialized tasks . they integrate LLMs with expert annotators to observe the impact of LLM suggestions .
Outcome:	The proposed model improves task completion speed but introduces anchoring bias . the proposed model is not suitable for open-ended analysis, but is capable of handling specialized tasks.

Copied to clipboard

Chenglin Wang, Yucheng Zhou, Shuang Chen, Tao Wang, Kai Zhang

Visual Data in NLP

Challenge:	Existing methods for inference are expensive and lack spatial redundancy . Discrete Diffusion Language Models are a promising paradigm for multimodal generation .
Approach:	They propose a locality-aware dynamic rescue method that exploits spatial Markov property of images.
Outcome:	The proposed method achieves an approximate 4 speedup over baselines on four text-to-image generation benchmarks.

Token-level Inference-Time Alignment for Vision-Language Models (2026.findings-acl)

Copied to clipboard

Kejia Chen, Junjun Zheng, Jiawen Zhang, Manxi Lin, Xiao Pan, Jiacong Hu, Jian Lou, Zunlei Feng, Mingli Song

Ranking Unraveled: Recipes for LLM Rankings in Head-to-Head AI Combat (2025.acl-long)

Challenge:	Vision-Language Models (VLMs) often prioritize linguistic fluency over visual fidelity . despite widespread adoption, VLMs often exhibit a critical failure mode: hallucination .
Approach:	They propose a framework for Token-level Inference-Time Alignment that steers the decoding process without updating the base model parameters.
Outcome:	The proposed framework improves performance on 13 benchmarks across architectures . it boosts LLaVA-1.5-7B by 8.6% on MMVet and achieves a 74.0 MMStar score .

Copied to clipboard

Roland Daynauth, Christopher Clarke, Krisztian Flautner, Lingjia Tang, Jason Mars

KARPA: A Training-free Method of Adapting Knowledge Graph as References for Large Language Model’s Reasoning Path Aggregation (2025.findings-acl)

Challenge:	Evaluating large language models (LLMs) is a complex task. Pairwise ranking has emerged as state-of-the-art method to evaluate human preferences.
Approach:	They propose to use pairwise ranking to evaluate human preferences . they propose to evaluate the robustness of ranking algorithms in LLMs .
Outcome:	The proposed methods are based on the principles of effective ranking and the robustness of several ranking algorithms in the context of LLMs.

Copied to clipboard

Siyuan Fang, Kaijing Ma, Tianyu Zheng, Xeron Du, Ningxuan Lu, Ge Zhang, Qingkun Tang

HATA: Trainable and Hardware-Efficient Hash-Aware Top-k Attention for Scalable Large Model Inference (2025.findings-acl)

Challenge:	Existing methods for large language models (LLMs) are limited by step-by-step decision-making on KGs, or require fine-tuning or pre-training on specific KG.
Approach:	They propose a framework that harnesses the global planning abilities of large language models (LLMs) for efficient and accurate KG reasoning.
Outcome:	Extensive experiments show that the proposed framework achieves state-of-the-art performance in KGQA tasks, delivering both high efficiency and accuracy.

Copied to clipboard

Ping Gong, Jiawei Yi, Shengnan Wang, Juncheng Zhang, Zewen Jin, Ouxiang Zhou, Ruibo Liu, Guanbin Xu, Youhui Bai, Bowen Ye, Kun Yuan, Tong Yang, Gong Zhang, Renhai Chen, Feng Wu, Cheng Li

SmartTrim: Adaptive Tokens and Attention Pruning for Efficient Vision-Language Models (2024.lrec-main)

Challenge:	Existing top-k attention methods struggle to strike a balance between efficiency and accuracy.
Approach:	They propose a top-k attention approach that integrates low-overhead techniques into the Top-k Attention process to achieve 7.2 speedup compared to vanilla full attention.
Outcome:	The proposed approach achieves 7.2 speedup compared to current top-k attention methods while maintaining model accuracy.

Copied to clipboard

Zekun Wang, Jingchang Chen, Wangchunshu Zhou, Haichao Zhu, Jiafeng Liang, Liping Shan, Ming Liu, Dongliang Xu, Qing Yang, Bing Qin

Summary Factual Inconsistency Detection Based on LLMs Enhanced by Universal Information Extraction (2025.findings-acl)

Challenge:	Experimental results show that SmartTrim accelerates the original model by 2-3 times with minimal performance degradation.
Approach:	They propose an adaptive acceleration framework which prunes redundant token representations and attention heads within each layer of the original model.
Outcome:	The proposed framework accelerates the original model by 2-3 times with minimal performance degradation across vision-language tasks.

Copied to clipboard

Anguo Li, Lei Yu

Beyond Chunks and Graphs: Retrieval-Augmented Generation through Triplet-Driven Thinking (2026.findings-acl)

Challenge:	Recent studies have shown that Large language models can detect factual inconsistencies in summaries but they lack the efficiency and explainability needed to be effective.
Approach:	They propose to decouple LLMs’ information extraction and reasoning capabilities to address key challenges and propose a framework for UIEFID to guide fine-tuned LLM methods in extracting unified structured information from documents and summaries.
Outcome:	The proposed framework improves the detection accuracy and reduces redundant reasoning on the AGGREFACT benchmark.

Copied to clipboard

Shengbo Gong, Xianfeng Tang, Qi He, Carl Yang, Wei Jin

When Should Dense Retrievers Be Updated in Evolving Corpora? Detecting Out-of-Distribution Corpora Using GradNormIR (2025.findings-acl)

Challenge:	Retrieval-augmented generation (RAG) is critical for reducing hallucinations and incorporating external knowledge into Large Language Models (LLMs).
Approach:	They propose a framework that leverages an LLM to decompose questions into searchable triplets with placeholders.
Outcome:	Empirical results show that T2RAG outperforms state-of-the-art multi-round and Graph RAG methods while reducing retrieval costs by up to 45%.

Copied to clipboard

Dayoon Ko, Jinyoung Kim, Sohyeon Kim, Jinhyuk Kim, Jaehoon Lee, Seonghak Song, Minyoung Lee, Gunhee Kim

Arctic-Text2SQL-R1: Simple Rewards, Strong Reasoning in Text-to-SQL (2026.findings-acl)

Challenge:	Dense retrievers encode text into embeddings to retrieve relevant documents . however, real-world corpora evolve, resulting in degraded retrieval performance . identifying when a dense retriever requires an update is critical for robust retrieval systems .
Approach:	They propose a task of predicting whether a corpus is out-of-distribution (OOD) relative to a dense retriever before indexing.
Outcome:	The proposed method detects whether a corpus is out-of-distribution (OOD) relative to a dense retriever before indexing.

Copied to clipboard

Zhewei Yao, Guoheng Sun, Łukasz Borchmann, Zheyu Shen, Minghang Deng, Bohan Zhai, Hao Zhang, Ang Li, Yuxiong He

GraphCheck: Multipath Fact-Checking with Entity-Relationship Graphs (2025.findings-emnlp)

Challenge:	Translating natural language questions into SQL is a core challenge in natural language understanding and human-computer interaction.
Approach:	They propose a reinforcement learning framework and model family to generate accurate, executable SQL using a lightweight reward signal based solely on execution correctness.
Outcome:	The proposed framework outperforms previous versions of 70B-class systems and achieves state-of-the-art execution accuracy across six diverse Text2SQL benchmarks.

Copied to clipboard

Hyewon Jeon, Jay-Yoon Lee

Gödel Agent: A Self-Referential Agent Framework for Recursively Self-Improvement (2025.acl-long)

Challenge:	GraphCheck is a framework for fact-checking complex claims that require multi-hop reasoning . Graphcheck excels in complex scenarios, but may be unnecessarily elaborate for simpler claims .
Approach:	They propose a framework that transforms claims into entity-relationship graphs for fact-checking . DP-GraphCheck employs a lightweight strategy selector to choose between direct prompting and GraphCheck adaptively.
Outcome:	The proposed framework outperforms existing methods in verification accuracy while achieving strong computational efficiency.

Copied to clipboard

Xunjian Yin, Xinyi Wang, Liangming Pan, Li Lin, Xiaojun Wan, William Yang Wang

HydraOpt: Navigating the Efficiency-Performance Trade-off of Adapter Merging (2025.emnlp-main)

Challenge:	Existing agentic systems cannot search the whole design space due to the restriction of human-designed components.
Approach:	They propose a Gödel Agent framework that allows agents to recursively improve themselves without relying on fixed algorithms or fixed algorithms.
Outcome:	The proposed framework surpasses manual crafted agents in performance, efficiency, and generalizability.

Copied to clipboard

Taha Ceritli, Ondrej Bohdal, Mete Ozay, Jijoong Moon, Kyenghun Lee, Hyeonmok Ko, Umberto Michieli

Injecting Domain-Specific Knowledge into Large Language Models: A Comprehensive Survey (2025.findings-emnlp)

Challenge:	Existing methods that produce a fixed trade-off between storage size and performance are often ineffective due to the growing size of large language models.
Approach:	They propose a model merging technique that capitalizes on similarities between low-rank adapters to reduce storage costs and improve performance.
Outcome:	The proposed method significantly reduces storage size (48% reduction) while outperforms existing merging techniques in terms of performance (0.2-1.8% drop).

Copied to clipboard

Zirui Song, Bin Yan, Yuhan Liu, Miao Fang, Mingzhe Li, Rui Yan, Xiuying Chen

Learn to Memorize: Scalable Continual Learning in Semiparametric Models with Mixture-of-Neighbors Induction Memory (2025.acl-long)

Challenge:	specialized LLMs are often limited in domain-specific applications that require specialized knowledge.
Approach:	They provide a comprehensive overview of four key methods to enhance large language models by integrating domain-specific knowledge.
Outcome:	The proposed methods are categorized into four key approaches: dynamic knowledge injection, static knowledge embedding, modular adapters, and prompt optimization.

Copied to clipboard

Guangyue Peng, Tao Ge, Wen Luo, Wei Li, Houfeng Wang

TALON: A Multi-Agent Framework for Long-Table Exploration and Question Answering (2025.emnlp-main)

Challenge:	Semiparametric language models (LMs) use static storage, which lacks learning capability and is disconnected from the internal information flow of the parametric models.
Approach:	They reconceptualize the non-parametric memory represented by kNN-LM as a learnable Mixture-of-Neighbors Induction Memory (MoNIM) this synergizes the induction capabilities of attention heads with the memorization strength of feed-forward networks .
Outcome:	The proposed model is a learnable Mixture-of-neighbors induction memory (MoNIM) it synergizes the induction capabilities of attention heads with the memorization strength of feed-forward networks (FFNs).

Copied to clipboard

Ruochun Jin, Xiyue Wang, Dong Wang, Haoqi Zheng, Yunpeng Qi, Silin Yang, Meng Zhang

Bypass Back-propagation: Optimization-based Structural Pruning for Large Language Models via Policy Gradient (2025.acl-long)

Challenge:	Existing approaches to query-relevant content retrieval fail to retrieve contextually relevant data.
Approach:	They propose a multi-agent framework for table question answering over long tables . TALON features a planning agent that iteratively invokes a tool agent to access tabular data .
Outcome:	The proposed framework achieves average accuracy improvements of 7.5% and 12.0% across all language models.

Copied to clipboard

Yuan Gao, Zujing Liu, Weizhong Zhang, Bo Du, Gui-Song Xia

From Domains to Instances: Dual-Granularity Data Synthesis for LLM Unlearning (2026.findings-acl)

Challenge:	Recent pruning methods rely on heuristically hand-crafted metrics, leading to suboptimal performance.
Approach:	They propose a method that optimizes pruning masks by minimizing back-propagation . they learn an underlying Bernoulli distribution to sample binary pruning mask samples .
Outcome:	The proposed method is able to support global and heterogeneous pruning without back-propagation.

Copied to clipboard

Xiaoyu Xu, Minxin Du, Zitong LI, Zi Liang, Zhibiao Guo, Zhang Shiyu, Peizhao Hu, Qingqing Ye, Haibo Hu

Challenge:	Currently, the evaluation of unlearning is limited due to the lack of granularity in the model.
Approach:	They propose a framework for synthesizing high-quality forget sets that exploits the target model per se to elicit data that matches its internal knowledge distribution through seed-guided and adversarial prompting.
Outcome:	The proposed framework achieves a superior balance of relevance, diversity, and efficiency across benchmarks.

ERRV: Eliciting Efficient Reasoning through Reasoning Vectors for Policy Optimization in Large Language Models (2026.findings-acl)

Copied to clipboard

Zhuowen Han, Lei Yang, Renren Jin, Dan Shi, Chenxi Sun, Deyi Xiong

Multi-view-guided Passage Reranking with Large Language Models (2025.emnlp-main)

Challenge:	Existing efforts to improve reasoning efficiency of large language models focus on modifying the reinforcement learning reward, such as adding length penalties.
Approach:	They propose a training framework that elicits efficient reasoning through reasoning vectors and a framework that allows the model to generate high-quality responses during reinforcement learning.
Outcome:	The proposed framework reduces reasoning length by 30% while maintaining stability, while retaining high accuracy.

Copied to clipboard

Jeongwoo Na, Jun Kwon, Eunseong Choi, Jongwuk Lee

Web Sitemap Knowledge Can Enhance Autonomous Browsing (2026.findings-acl)

Challenge:	Existing models rely on autoregressive generation and sliding window strategies to rank passages, which incur heavy computational overhead as the number of passages increases.
Approach:	They propose a non-generative LLM-based reranking method that encodes query-passage information into diverse view embeddings without being influenced by external biases.
Outcome:	The proposed model matches the performance of much larger 7B-scale fine-tuned models while achieving a 100x reduction in inference latency.

Copied to clipboard

Yuyao Zhang, Hongyu Lu, Jiajie Jin, Hongjin Qian, Shiyu Li, Zhao Yang, Yutao Zhu, Ji-Rong Wen, Zhicheng Dou

A Parameter-Efficient and Fine-Grained Prompt Learning for Vision-Language Models (2025.acl-long)

Challenge:	Existing web agents suffer from limited robustness, efficiency and task success due to lack of structural understanding of websites and lack of browsing priors in pre-trained models.
Approach:	They propose an agent-oriented sitemap protocol that integrates structured website knowledge into web agents.
Outcome:	The proposed agent-oriented sitemap improves robustness, efficiency and effectiveness without extra training.

Copied to clipboard

Yongbin Guo, Shuzhen Li, Zhulin Liu, Tong Zhang, C.L.Philip Chen

KGE Calibrator: An Efficient Probability Calibration Method of Knowledge Graph Embedding Models for Trustworthy Link Prediction (2025.emnlp-main)

Challenge:	Current vision-language models extract semantic information from large-scale cross-modal associations, limiting performance and efficiency.
Approach:	They propose a detail-oriented prompt learning method to implement fine-grained multi-modal semantic alignment with merely 0.25M trainable parameters.
Outcome:	The proposed method implements fine-grained multi-modal semantic alignment with merely 0.25M trainable parameters.

Copied to clipboard

Yang Yang, Mohan Timilsina, Edward Curry

Clustered Self-Assessment: A Simple yet Effective Method for Uncertainty Quantification in Large Language Models (2026.findings-acl)

Challenge:	Existing methods for probability calibration of knowledge graph embedding models are ill-suited for KGEs.
Approach:	They propose a method to calibrate knowledge graph embedding models for ranking-based link prediction using a Jump Selection Strategy and Multi-Binning Scaling to enhance reliability.
Outcome:	Experiments show that the KGEC outperforms existing calibration methods in terms of effectiveness and efficiency.

Copied to clipboard

Qi Cao, Takeshi Kojima, Andrew Gambardella, Helinyi Peng, Yutaka Matsuo, Yusuke Iwasawa

Accelerated Test-Time Scaling with Model-Free Speculative Sampling (2025.emnlp-main)

Challenge:	Existing methods for uncertainty quantification in large language models rely on indirect signals, such as entropy across sampled generations, which can be difficult to interpret and do not fully leverage the model’s ability to assess its own uncertainty.
Approach:	They propose a method that groups sampled generations into semantically distinct clusters and uses the probability assigned by the LLM to each option as a confidence estimate.
Outcome:	The proposed method outperforms baseline methods and achieves competitive performance with as few as two additional samples.

Copied to clipboard

Woomin Song, Saket Dingliwal, Sai Muralidhar Jayanthi, Bhavana Ganesh, Jinwoo Shin, Aram Galstyan, Sravan Babu Bodapati

MiniRAG: A Lightweight RAG system with Small Language Models (2026.acl-long)

Challenge:	Language models have demonstrated remarkable capabilities in reasoning tasks through test-time scaling techniques like best-of-N sampling and tree search.
Approach:	They propose a model-free speculative decoding approach that exploits redundancy in reasoning trajectories to achieve significant acceleration without compromising accuracy.
Outcome:	The proposed approach reduces inference latency by 60-65% while maintaining accuracy.

Copied to clipboard

Tianyu Fan, Jingyuan Wang, Xubin Ren, Chao Huang

AOT*: Efficient Synthesis Planning via LLM-Empowered AND-OR Tree Search (2026.findings-acl)

Challenge:	Existing RAG frameworks rely on Large Language Models (LLMs) for all stages of the process, resulting in high computational costs and resource demands.
Approach:	They propose a semantic-aware heterogeneous graph indexing mechanism that combines text chunks and named entities in a unified structure and a lightweight topology-enhanced retrieval approach that leverages graph structures for efficient knowledge discovery without requiring advanced language capabilities.
Outcome:	The proposed system achieves comparable performance to LLM-based methods while requiring only 25% of the storage space.

Copied to clipboard

Xiaozhuang Song, Xuanhao Pan, Xinjian Zhao, Hangting Ye, Shufei Zhang, Jian Tang, Tianshu Yu

Beyond Single Representations: Multi-Model Embedding Fusion for Stable Text Classification (2026.acl-long)

Challenge:	Experimental evaluation shows that AOT* achieves competitive solve rates using 3-5 fewer iterations than existing LLM-based approaches.
Approach:	They propose a framework that integrates LLM-generated chemical synthesis pathways with systematic AND-OR tree search.
Outcome:	Experimental results show that AOT* improves search efficiency and solves faster than existing approaches.

Copied to clipboard

Jiho Gwak, Yuchul Jung

SAFER: Advancing Safety Alignment via Efficient Ex-Ante Reasoning (2026.findings-acl)

Challenge:	Existing studies on embedding fusion have not evaluated the effectiveness of individual layers or the impact of combining embeddables from multiple models.
Approach:	They propose to combine embeddings from multiple models to improve performance across NLP tasks.
Outcome:	The proposed method improves performance on low-resource datasets and reduces the impact of any single model as the number of integrated models increases.

Copied to clipboard

Kehua Feng, Keyan Ding, Yuhao Wang, Menghan Li, Fanjunduo Wei, Xinda Wang, Huajun Chen

LiCoMemory: Lightweight and Cognitive Agentic Memory for Efficient Long-Term Reasoning (2026.findings-acl)

Challenge:	Existing alignment methods struggle to cover diverse safety scenarios and remain vulnerable to adversarial attacks.
Approach:	They propose a framework for 'Safety' alignment via eFficient' Ex-Ante-R**easoning that instantiates structured Ex-Ance reasoning and embeds predefined safety rules.
Outcome:	The proposed framework enhances safety performance while maintaining usefulness and efficiency.

Copied to clipboard

Zhengjun Huang, Zhoujin Tian, Qintian Guo, Fangyuan Zhang, Yingli Zhou, Di Jiang, Zeying Xie, Xiaofang Zhou

From Individual Excellence to Collective Sustainability: Seeking Strategic Equilibrium in Proactive Multi-Agent Teams (2026.findings-acl)

Challenge:	Large Language Models are constrained by limited context windows and lack of persistent memory . recent efforts address these limitations via external memory architectures .
Approach:	They propose an end-to-end agentic memory framework for real-time updating and retrieval that integrates hierarchical and temporal indexing layers.
Outcome:	The proposed framework outperforms established benchmarks in temporal reasoning, multi-session consistency, and retrieval efficiency.

Copied to clipboard

Tong Zhang, Yang Wu, Yufei Shi, Rujing Yao, Zhuoren Jiang, Xiaozhong Liu

Uncertainty-Aware Test-Time Search for Optimization Problem Solving (2026.acl-long)

Challenge:	a team of proactive agents suffer from a greedy optimization for immediate task accuracy . a new approach to improve team collaboration is based on the opportunity cost .
Approach:	They propose a game-theoretic proactive multi-agent reinforcement learning framework to solve this imbalance . they use a Positive-Unlabeled scorer to anchor intervention quality under sparse supervision .
Outcome:	The proposed framework maintains high performance while preventing experts from over-developing.

Copied to clipboard

Linlin Yu, Xujiang Zhao, Dong Li, Yanchi Liu, Wei Cheng, Zhengzhang Chen, Chen Zhao, Feng Chen, Haifeng Chen

EVOTOOL: Self-Evolving Tool-Use Policy Optimization in LLM Agents via Blame-Aware Mutation and Diversity-Aware Selection (2026.acl-long)

Challenge:	Language model hallucinations and limited availability of labeled datasets often result in misaligned formulations, code errors and feasibility failures.
Approach:	They propose a Monte Carlo Tree Search framework that automates optimization problems from natural language descriptions with efficiency and reliability.
Outcome:	The proposed framework achieves state-of-the-art solution accuracy and reduces token usage.

Copied to clipboard

Shuo Yang, Caren Han, Xueqi Ma, Yan Li, Mohammad Reza Ghasemi Madani, Eduard Hovy

Challenge:	Existing approaches to optimize tool-use policies are monolithic and prone to entangling behaviors.
Approach:	They propose a framework that decomposes agent’stool-use policy into four modules and improves them via three mechanisms.
Outcome:	The proposed framework outperforms strong baselines on bothGPT-4.1 and Qwen3-8B while maintaining superior efficiency and transferability.

AIPO: Adaptive Information Guided Token-Level Reinforcement Learning for Large Language Model Reasoning (2026.acl-long)

Copied to clipboard

Bin Chen, Hongfei Ye, Huiyang Wang, Wenxi Liu, Yu Zhang, Furui Liu

MorphBPE: Morphology-Aware Tokenization for Efficient LLM Training (2026.findings-acl)

Challenge:	Existing RLVR methods focus on all generated tokens rather than on which tokens contribute to reasoning.
Approach:	They propose to use a Random–Fourier approximation of the Hilbert–Schmidt Independence Criterion to focus updates on decisive tokens discovered on the fly to improve the efficiency of mutual-information estimation.
Outcome:	The proposed approach yields +20% accuracy over strong RLVR baselines while updating merely 10% of tokens, demonstrating superior efficiency and effectiveness.

Copied to clipboard

Ehsaneddin Asgari, Yassine El Kheir, MohammadAli SadraeiJavaheri, Ali Nazari

PolitNuggets: Benchmarking Agentic Discovery of Long-Tail Political Facts (2026.acl-long)

Challenge:	Tokenization is a key design choice in modern NLP systems and a critical bottleneck for multilingual Large Language Models.
Approach:	They propose a tokenization extension that constrains merge operations to respect morpheme boundaries while preserving inference.
Outcome:	The proposed tokenization improves morphological coherence and language model cross-entropy in four languages.

Copied to clipboard

Yifei Zhu

Revealing the Seen, Imagining the Beyond: A Survey of Image-Grounded Chain-of-Thought Reasoning in Multimodal LLMs (2026.acl-long)

Challenge:	Large Reasoning Models (LRMs) are embedded in agentic frameworks and are under-evaluated.
Approach:	They propose a multilingual benchmark for agentic information synthesis using PolitNuggets . they standardize evaluation with an optimized Supervisor–Searcher multi-agent system .
Outcome:	The proposed model can discover and synthesize "long-tail" facts from dispersed sources.

Copied to clipboard

Qihua Dong, Yitian Zhang, Huimin Zeng, Yizhou Wang, Jianglin Lu, Kuo Yang, Yun Fu

DyBBT: Dynamic Balance via Bandit-inspired Targeting for Dialog Policy with Cognitive Dual Systems (2026.acl-long)

Challenge:	Recent advances in Multimodal Large Language Models (MLLMs) have shifted visual reasoning from tool-calling to end-to-end perceptionreasoning.
Approach:	They synthesize the emerging paradigm of Image-Grounded Chain-of-Thought (IG-CoT) they propose a method-centric taxonomy covering prompting, supervised fine-tuning, and reinforcement learning .
Outcome:	The proposed model is based on a method-centric taxonomy and benchmarks.

Copied to clipboard

Shuyu Zhang, Yifan Wei, Jialuo Yuan, Xinru Wang, Yanmin Zhu, Yujie Liu, Bin Li