Papers by Song Sun

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

Challenge:	Existing benchmarks focus more on end-to-end performance, but neglect the underlying principles of knowledge acquisition and generalization.
Approach:	They propose a benchmark specifically designed to explore the problem-solving principles by decomposing 6.5K visual math problems into 10.9K step-level questions for evaluation.
Outcome:	The proposed benchmark covers 6.5K visual math problems and 10.9K step-level questions spanning 5 layers of knowledge granularity and 67 hierarchical knowledge concepts.

Copied to clipboard

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

TR-Rules: Rule-based Model for Link Forecasting on Temporal Knowledge Graph Considering Temporal Redundancy (2023.findings-emnlp)

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

Ningyuan Li, Haihong E, Shi Li, Mingzhi Sun, Tianyu Yao, Meina Song, Yong Wang, Haoran Luo

ATLANTIS: Weak-to-Strong Learning via Importance Sampling (2025.acl-long)

Challenge:	Existing models suffer from temporal redundancy when leveraged under dynamic settings.
Approach:	They propose a temporal knowledge graph extrapolation method which solves temporal redundancy issues by using cyclic rules to capture more information lurking in TKGs.
Outcome:	The proposed model captures more information lurking in TKGs, and also mines and properly leverages acyclic rules, which has not been explored by existing models.

Copied to clipboard

Yi Liu, Guoyin Wang, Shicheng Li, Feifan Song, Xu Sun

Challenge:	ATLANTIS is a new technique to improve the performance of large language models.
Approach:	They propose a new technique to bridge the gap between the distribution of current datasets and the real-world data distribution by using importance sampling.
Outcome:	The proposed technique can bring consistent and significant improvements to models’ performance and can be flexibly transferred among models with different structures.

An Empirical Study of Iterative Refinements for Non-autoregressive Translation (2025.acl-long)

Copied to clipboard

Yisheng Xiao, Pei Guo, Zechen Sun, Juntao Li, Kai Song, Min Zhang

STICKERCONV: Generating Multimodal Empathetic Responses from Scratch (2024.acl-long)

Challenge:	Iterative non-autoregressive (NAR) models have recently demonstrated impressive performance in varied generation tasks, surpassing the autoregressive Transformer.
Approach:	They propose a strategy to conduct efficient refinements without performance declines by using two simple metrics to identify potential problems existing in current refinement processes.
Outcome:	The proposed model outperforms the autoregressive Transformer by around one BLEU on average.

Copied to clipboard

Yiqun Zhang, Fanheng Kong, Peidong Wang, Shuang Sun, SWangLing SWangLing, Shi Feng, Daling Wang, Yifei Zhang, Kaisong Song

V-ALPHASOCIAL: Benchmark and Self-Reflective Chain-of-Thought Generation for Visual Social Commonsense Reasoning (2025.findings-acl)

Challenge:	Prior studies on stickers focused on sentiment analysis and recommendation systems, overlooking their vast potential in empathetic response generation.
Approach:	They propose a multimodal empathetic dialogue dataset, STICKERCONV, which simulates human behavior with stickers, and propose evaluative metrics based on LLM.
Outcome:	The proposed framework generates contextually relevant and emotionally resonant multimodal empathetic responses, contributing to the advancement of more nuanced and engaging e-dialog systems.

Copied to clipboard

Zongyu Lin, Zhikun Xu, Xiaohan Song, Yixin Wan, Xingcheng Yao, Tsung-Han Lin, Selina Song, Pranav Subbaraman, Ben Zhou, Kai-Wei Chang, Yizhou Sun

AdDriftBench: A Benchmark for Detecting Data Drift and Label Drift in Short Video Advertising (2025.findings-emnlp)

Challenge:	Social commonsense reasoning is a multimodal task that requires both textual and visual cues.
Approach:	They propose a method that integrates visual cues into social commonsense reasoning tasks.
Outcome:	The proposed method improves social commonsense reasoning on a multimodal foundation model.

Copied to clipboard

Yinghao Song, Xiangji Zeng, Shuai Cui, Lu Sun, Zhaowei Liu, Yuan Yuan, Yulu Wang, Hai Zhou, Zhaohan Gong

BSharedRAG: Backbone Shared Retrieval-Augmented Generation for the E-commerce Domain (2024.findings-emnlp)

Challenge:	Short video advertising scenarios present unique challenges due to data drift (DD) and label drift (LD).
Approach:	They propose to use data drift and label drift to evaluate models under rapidly shifting content distributions and labeling scenarios to assess their generalization capabilities.
Outcome:	The proposed model performs moderately in short video advertising contexts, particularly in handling fine-grained semantics and adapting to shifting instructions.

Copied to clipboard

Kaisi Guan, Qian Cao, Yuchong Sun, Xiting Wang, Ruihua Song

Q-TOD: A Query-driven Task-oriented Dialogue System (2022.emnlp-main)

Challenge:	Existing work adopts separate modules for retrieval and generation, which may be suboptimal since the retrieval task and generation task cannot benefit from each other to improve performance.
Approach:	They propose a backbone-shared RAG framework that uses a domain-specific corpus to continuously pre-train a model and then trains two plug-and-play Low-Rank Adaptation modules based on the shared backbone to minimize retrieval and generation losses respectively.
Outcome:	The proposed framework outperforms baseline models by 5% and 13% in Hit@3 upon two datasets in retrieval evaluation and by 23% in terms of BLEU-3 in generation evaluation.

Copied to clipboard

Xin Tian, Yingzhan Lin, Mengfei Song, Siqi Bao, Fan Wang, Huang He, Shuqi Sun, Hua Wu

MAGID: An Automated Pipeline for Generating Synthetic Multi-modal Datasets (2024.naacl-long)

Challenge:	Existing pipelined task-oriented dialogue systems have difficulties adapting to unseen domains . end-to-end systems are plagued by large-scale knowledge bases in practice .
Approach:	They propose a query-driven task-oriented dialogue system that extracts dialogue context information into a natural language query.
Outcome:	The proposed system outperforms strong baselines and establishes a new state-of-the-art performance on three publicly available task-oriented dialogue datasets.

Copied to clipboard

Hossein Aboutalebi, Hwanjun Song, Yusheng Xie, Arshit Gupta, Lijia Sun, Hang Su, Igor Shalyminov, Nikolaos Pappas, Siffi Singh, Saab Mansour

Noisy Multi-Label Text Classification via Instance-Label Pair Correction (2024.findings-naacl)

Challenge:	Existing approaches to augment textual dialogues with retrieved images pose privacy, diversity, and quality constraints.
Approach:	They propose a framework to augment text-only dialogues with diverse and high-quality images by using a diffusion model and a feedback loop.
Outcome:	The proposed framework is comparable to or better than baselines, with significant improvements in human evaluation, especially against retrieval baselines where the image database is small.

Copied to clipboard

Pengyu Xu, Mingyang Song, Linkaida Liu, Bing Liu, Hongjian Sun, Liping Jing, Jian Yu

Information Extraction & Text Mining

Challenge:	Noise is a significant challenge for machine learning models, especially deep learning models.
Approach:	They propose a holistic selection metric that identifies noisy pairs while considering global loss information and instance-specific ranking information.
Outcome:	The proposed approach significantly improves performance in noisy multi-label text classification tasks.

Both Text and Images Leaked! A Systematic Analysis of Data Contamination in Multimodal LLM (2025.findings-emnlp)

Copied to clipboard

Dingjie Song, Sicheng Lai, Mingxuan Wang, Shunian Chen, Lichao Sun, Benyou Wang

Progressive Self-Supervised Attention Learning for Aspect-Level Sentiment Analysis (P19-1)

Challenge:	Existing methods for unimodal large language models are inadequate for MLLMs due to multimodal data complexity and multi-phase training.
Approach:	MM-DETECT analyzes data contamination using a framework that defines two contamination categories - unimodal and cross-modal .
Outcome:	The proposed framework quantifies contamination severity across multiple-choice and caption-based Visual Question Answering tasks.

Copied to clipboard

Jialong Tang, Ziyao Lu, Jinsong Su, Yubin Ge, Linfeng Song, Le Sun, Jiebo Luo

Sentiment Analysis

Challenge:	Experimental results show that our proposed approach yields better attention mechanisms . dominant ASC models are mostly discriminative classifiers based on manual feature engineering .
Approach:	They propose a self-supervised approach to aspect-level sentiment classification that mines useful attention supervision information from a training corpus to refine attention mechanisms.
Outcome:	The proposed approach yields better attention mechanisms on multiple datasets.

OpenEval: Benchmarking Chinese LLMs across Capability, Alignment and Safety (2024.acl-demos)

Copied to clipboard

Chuang Liu, Linhao Yu, Jiaxuan Li, Renren Jin, Yufei Huang, Ling Shi, Junhui Zhang, Xinmeng Ji, Tingting Cui, Liutao Liutao, Jinwang Song, Hongying Zan, Sun Li, Deyi Xiong

D-Artemis: A Deliberative Cognitive Framework for Mobile GUI Multi-Agents (2026.findings-acl)

Challenge:	a rapid development of Chinese large language models poses big challenges for efficient LLM evaluation.
Approach:	They propose an evaluation testbed that benchmarks Chinese LLMs across capability, alignment and safety.
Outcome:	The evaluation platform OpenEval benchmarks Chinese LLMs across capability, alignment and safety.

Copied to clipboard

Hongze Mi, Yibo Feng, WenJie Lu, Yuqi Wang, Jinyuan Li, Song Cao, He Cui, Tengfei Tian, Xuelin Zhang, Haotian Luo, Di Sun, Jun Fang, Hua Chai, Naiqiang Tan, Gang Pan

Challenge:	Graphical User Interface (GUI) agents aim to automate a wide spectrum of human tasks by emulating user interaction.
Approach:	They propose a deliberative framework that leverages a fine-grained tip retrieval mechanism to inform its decision-making process.
Outcome:	The proposed framework achieves SOTA among open-source general models on AndroidWorld and ScreenSpot-V2 . it leverages a fine-grained, app-specific tip retrieval mechanism to inform its decision-making process .

Improving Meta-learning for Low-resource Text Classification and Generation via Memory Imitation (2022.acl-long)

Copied to clipboard

Yingxiu Zhao, Zhiliang Tian, Huaxiu Yao, Yinhe Zheng, Dongkyu Lee, Yiping Song, Jian Sun, Nevin Zhang

Information Extraction & Text Mining

Challenge:	Building models of natural language processing (NLP) is challenging in low-resource scenarios where limited data are available.
Approach:	They propose a memory imitation meta-learning method that enhances the model’s reliance on support sets for task adaptation.
Outcome:	The proposed method outperforms baselines on both text classification and generation tasks.

Joint Semantic and Strategy Matching for Persuasive Dialogue (2023.findings-emnlp)

Copied to clipboard

Chuhao Jin, Yutao Zhu, Lingzhen Kong, Shijie Li, Xiao Zhang, Ruihua Song, Xu Chen, Huan Chen, Yuchong Sun, Yu Chen, Jun Xu

VQA-Augmented Machine Translation with Cross-Modal Contrastive Learning (2025.findings-emnlp)

Challenge:	Persuasive dialogue models rely on utterance semantic matching and a key aspect has been ignored . compared with utterrance semantics, conversation strategies are high-level concepts, which can be informative and provide complementary information to achieve effective persuation.
Approach:	They propose to model conversation semantics and strategies to match them using a BERT-like module and an auto-regressive predictor.
Outcome:	The proposed model improves state-of-the-art by 5% on a small and 37% on 'large' datasets.

Copied to clipboard

Zhihui Zhang, Shiliang Sun, Jing Zhao, Tengfei Song, Hao Yang

Reasoning-Enhanced Domain-Adaptive Pretraining of Multimodal Large Language Models for Short Video Content Governance (2025.emnlp-industry)

Challenge:	Existing multimodal machine translation methods often extract visual features using pre-trained models while learning text features from scratch, leading to representation imbalance.
Approach:	They propose a cross-modal VQA-augmented multimodal machine translation method . it aligns image-source text pairs and image-question text pairs through dual-text contrastive learning .
Outcome:	The proposed method outperforms state-of-the-art methods on multiple evaluation metrics.

Copied to clipboard

Zixuan Wang, Yu Sun, Hongwei Wang, Baoyu Jing, Xiang Shen, Xin Dong, Zhuolin Hao, Hongyu Xiong, Yang Song

RPC-Bench: A Fine-grained Benchmark for Research Paper Comprehension (2026.acl-long)

Challenge:	Existing approaches to identifying inappropriate content require extensive human-labeled data and lack cross-issue generalization.
Approach:	They propose a reasoning-enhanced multimodal large language model (MLLM) pretraining paradigm for unified inappropriate content detection.
Outcome:	The proposed model improves the MLLM's performance in both zero-shot and supervised fine-tuning settings and shows strong generalization capabilities to emergent, previously unseen issues.

Copied to clipboard

Yelin Chen, Fanjin Zhang, Suping Sun, Yunhe Pang, Yuanchun Wang, Jian Song, XiaoYan Li, Lei Hou, Shu Zhao, Jie Tang, Juanzi Li

Challenge:	Existing benchmarks for understanding research papers offer limited fine-grained evaluation at scale.
Approach:	They propose a large-scale question-answering benchmark built from review–rebuttal exchanges of high-quality computer science papers.
Outcome:	The proposed model is based on human-verified QA pairs and contains 15K questions.

CRASpell: A Contextual Typo Robust Approach to Improve Chinese Spelling Correction (2022.findings-acl)

Copied to clipboard

Shulin Liu, Shengkang Song, Tianchi Yue, Tao Yang, Huihui Cai, TingHao Yu, Shengli Sun

Locally Differentially Private In-Context Learning (2024.lrec-main)

Challenge:	Recent research on Chinese spelling correction methods has poor performance on multi-typo texts.
Approach:	They propose to use Bert-based Chinese spelling correction models to overcome these limitations by constructing a noisy context for each training sample and a copy mechanism to encourage the model to choose the input character when the miscorrected and input character are both valid.
Outcome:	The proposed model outperforms state-of-the-art models on widely used benchmarks and achieves a remarkable gain.

Copied to clipboard

Chunyan Zheng, Keke Sun, Wenhao Zhao, Haibo Zhou, Lixing Jiang, Shaoyang Song, Chunlai Zhou

TofuEval: Evaluating Hallucinations of LLMs on Topic-Focused Dialogue Summarization (2024.naacl-long)

Challenge:	Large pretrained language models (LLMs) have shown surprising In-Context Learning ability.
Approach:	They propose a locally differentially private framework of in-context learning for LLMs that can be augmented with a private database for some specific task.
Outcome:	The proposed framework can predict labels without additional parameter modifications without input-label pairs .

Copied to clipboard

Liyan Tang, Igor Shalyminov, Amy Wong, Jon Burnsky, Jake Vincent, Yu’an Yang, Siffi Singh, Song Feng, Hwanjun Song, Hang Su, Lijia Sun, Yi Zhang, Saab Mansour, Kathleen McKeown

SOLAR: Serendipity Optimized Language Model Aligned for Recommendation (2025.findings-emnlp)

Challenge:	Existing LLMs hallucinate significant amounts of factual errors in the dialogue domain, regardless of the model’s size.
Approach:	They propose to evaluate topic-focused dialogue summarization by using large language models (LLMs) they use human annotations to evaluate factual consistency and explain factually inconsistent sentences.
Outcome:	The proposed evaluation benchmark on topic-focused dialogue summarization shows that existing LLMs hallucinate significant amounts of factual errors regardless of the model’s size.

Copied to clipboard

Zichen Yuan, Lifan Sun, Yucen Zhuang, Yue Wang, Xinyuan Song, Tianqi Xu, Siyuan Li, Junchen Fu, Youhua Li, Sirui Hong, Jiaqi Chen, Joemon M. Jose, Yongxin Ni

Human-in-the-loop Robotic Grasping Using BERT Scene Representation (2022.coling-1)

Challenge:	Large Language Models have shown strong potential in recommendation tasks . however, their application to serendipity-oriented recommendations remains challenging .
Approach:	They propose a domain-adaptive instruction tuning method that aligns Large Language Models with recommendation tasks.
Outcome:	The proposed framework bridges the domain gap between LLMs and recommendation tasks.

Copied to clipboard

Yaoxian Song, Penglei Sun, Pengfei Fang, Linyi Yang, Yanghua Xiao, Yue Zhang

MultiCodeAttack: Iterative Jailbreak Attacking on LLMs with Multi-Code Prompt Injection (2026.findings-acl)

Challenge:	Existing approaches for robotic grasping in cluttered scenes are expensive and lack structure information.
Approach:	They propose a human-in-the-loop framework for robotic grasping in cluttered scenes . they substitute scene-graph representation with a text representation of the scene using BERT .
Outcome:	The proposed framework outperforms object-agnostic and scene-graph based methods on robots and physical robots.

Copied to clipboard

Weifeng Sun, Meng Yan, Zhou Yang, Yuchen Chen, Song Sun, David Lo

Democratizing Reasoning Ability: Tailored Learning from Large Language Model (2023.emnlp-main)

Challenge:	Existing approaches to jailbreak rely on fixed template design and a single programming language . however, existing approaches do not consider language diversity or adaptive template evolution .
Approach:	They propose a structured jailbreak framework that explores and optimizes multi-language code templates.
Outcome:	The proposed framework outperforms existing jailbreak baselines and produces higher harmful outputs than baseline methods.

Copied to clipboard

Zhaoyang Wang, Shaohan Huang, Yuxuan Liu, Jiahai Wang, Minghui Song, Zihan Zhang, Haizhen Huang, Furu Wei, Weiwei Deng, Feng Sun, Qi Zhang

Auto Search Indexer for End-to-End Document Retrieval (2023.findings-emnlp)

Challenge:	Large language models (LLMs) exhibit impressive emergent abilities in natural language processing, but their democratization is hindered due to huge computation requirements and closed-source nature.
Approach:	They propose a tailored learning approach to distill the exclusive reasoning ability to smaller LMs to facilitate democratization.
Outcome:	The proposed approach enables the democratization of the exclusive reasoning ability by leveraging the black-box model as a reasoning teacher.

Copied to clipboard

Tianchi Yang, Minghui Song, Zihan Zhang, Haizhen Huang, Weiwei Deng, Feng Sun, Qi Zhang

Can Large Language Models Mine Interpretable Financial Factors More Effectively? A Neural-Symbolic Factor Mining Agent Model (2024.findings-acl)

Challenge:	Generative retrieval heavily relies on the “preprocessed” document identifiers, thus limiting its retrieval performance and ability to retrieve new documents.
Approach:	They propose a fully end-to-end retrieval paradigm that can learn the best docids for existing and new documents automatically via a semantic indexing module.
Outcome:	The proposed model outperforms baselines on public and industrial datasets and can handle new documents.

Copied to clipboard

Zhiwei Li, Ran Song, Caihong Sun, Wei Xu, Zhengtao Yu, Ji-Rong Wen

Using Customer Service Dialogues for Satisfaction Analysis with Context-Assisted Multiple Instance Learning (D19-1)

Challenge:	Existing factor mining models are inefficient and inefficient, resulting in a significant challenge to extract interpretable factors.
Approach:	They propose a model that integrates the strengths of both neural and symbolic models for factor mining.
Outcome:	The proposed model surpasses the SOTA RankIC and RankICIR in predicting S&P 500 returns on real-world stock market data.

Copied to clipboard

Kaisong Song, Lidong Bing, Wei Gao, Jun Lin, Lujun Zhao, Jiancheng Wang, Changlong Sun, Xiaozhong Liu, Qiong Zhang

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

Challenge:	Existing studies fail to provide comprehensive service satisfaction analysis . Existing models fail to include satisfaction polarity classification and sentimental utterance identification .
Approach:	They propose a model that predicts customer sentiments and aggregates them into service satisfaction polarity.
Outcome:	The proposed model predicts customer sentiments and aggregates them into service satisfaction polarity and reasoning clues.

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

Datasets for Scientific Literature Understanding: A Survey (2026.findings-acl)

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

Yuanzhe Zhang, Xun Zhao, Maodi Hu, Xi Sun, Donghuan Song, Zhixiong Zhang

Challenge:	Empowering machines to understand scientific literature is crucial for accelerating scientific discovery and advancing the AI for Science paradigm.
Approach:	They propose a systematic taxonomy that organizes resources spanning structural understanding, text understanding, multimodal understanding and pre-training/instruction fine-tuning.
Outcome:	The proposed taxonomy organizes resources spanning structural understanding, text understanding, multimodal understanding and pre-training/instruction fine-tuning.

MobileBERT: a Compact Task-Agnostic BERT for Resource-Limited Devices (2020.acl-main)

Copied to clipboard

Zhiqing Sun, Hongkun Yu, Xiaodan Song, Renjie Liu, Yiming Yang, Denny Zhou

Controllable Text Generation with Residual Memory Transformer (2024.findings-acl)

Challenge:	Empirical studies show that MobileBERT is 4.3x smaller and 5.5x faster than BERT_BASE . BERT is one of the largest models ever in NLP, but suffers from heavy model size and high latency .
Approach:	They propose a tool to compress and accelerate the popular BERT model by task-agnostic application.
Outcome:	The proposed model is 4.3x smaller and 5.5x faster than BERT_BASE . it achieves competitive results on well-known benchmarks .

Copied to clipboard

Hanqing Zhang, Si Sun, Haiming Wu, Dawei Song

Imagination and Contemplation: A Balanced Framework for Semantic-Augmented Multimodal Machine Translation (2025.findings-emnlp)

Challenge:	Large-scale Causal Language Models (CLMs) have been successful in text generation, but there is still a challenge to control the generation process.
Approach:	They propose a non-intrusive, lightweight control plugin to control the generation process of a CLM at arbitrary time steps.
Outcome:	The proposed plugin can handle any type of control conditions and cooperate with the base CLM through a residual learning paradigm.

Copied to clipboard

Zhuang Yu, Shiliang Sun, Jing Zhao, Tengfei Song, Hao Yang

Text-Tuple-Table: Towards Information Integration in Text-to-Table Generation via Global Tuple Extraction (2024.emnlp-main)

Challenge:	Multimodal Machine Translation (MMT) is effective in resolving linguistic ambiguities, but visual information often introduces redundancy or noise, potentially impairing translation quality.
Approach:	They propose a semantic-augmented framework that integrates "Imagination" and "Contemplation" they first generate synthetic images from source text and align them with authentic images via an optimal transport loss .
Outcome:	The proposed framework outperforms baselines on translation datasets with visually ambiguous or weakly correlated content.

Copied to clipboard

Zheye Deng, Chunkit Chan, Weiqi Wang, Yuxi Sun, Wei Fan, Tianshi Zheng, Yauwai Yim, Yangqiu Song

InMind: Evaluating LLMs in Capturing and Applying Individual Human Reasoning Styles (2025.emnlp-main)

Challenge:	Existing approaches to condensing textual information into concise and structured tables are limited in their applicability in broader contexts.
Approach:	They propose a benchmark dataset for generating summary tables of competitions based on real-time commentary texts that incorporates large-scale textual information into concise and structured tables.
Outcome:	The proposed method exhibits strong generalization abilities, surpassing previous approaches on several other text-to-table datasets.

Copied to clipboard

Zizhen Li, Chuanhao Li, Yibin Wang, Qi Chen, Diping Song, Yukang Feng, Jianwen Sun, Jiaxin Ai, Fanrui Zhang, Mingzhu Sun, Kaipeng Zhang

SynFix: Dependency-Aware Program Repair via RelationGraph Analysis (2025.findings-acl)

Challenge:	Recent large language models (LLMs) have demonstrated strong reasoning abilities across complex mathematical and scientific domains.
Approach:	They propose a framework to assess whether LLMs can capture and apply personalized reasoning styles in social deduction games.
Outcome:	The proposed framework evaluates LLMs on the game Avalon and shows that they can capture and apply individualized reasoning styles.

Copied to clipboard

Xunzhu Tang, Jiechao Gao, Jin Xu, Tiezhu Sun, Yewei Song, Saad Ezzini, Wendkûuni C. Ouédraogo, Jacques Klein, Tegawendé F. Bissyandé

Scale-VAE: Preventing Posterior Collapse in Variational Autoencoder (2024.lrec-main)

Challenge:	Existing methods for resolving repository-level debugging are limited by their interdependencies.
Approach:	They propose a RelationGraph-based approach that integrates large language models with structural search and synchronization techniques for coordinated program repair across codebases.
Outcome:	SynFix resolves 52.33% of issues in SWE-bench-lite, 55.8% in Swe-bech-verified and 29.86% in S WE-beach-full.

Copied to clipboard

Tianbao Song, Jingbo Sun, Xin Liu, Weiming Peng

CapArena: Benchmarking and Analyzing Detailed Image Captioning in the LLM Era (2025.findings-acl)

Challenge:	Variational autoencoder (VAE) is a widely used generative model . but when employing strong autoregressive generation networks, VAE tends to converge to a degenerate local optimum known as posterior collapse.
Approach:	They propose a model called Scale-VAE to solve a posterior collapse problem . they use a factor to keep the posterior dimension discriminative across data instances .
Outcome:	The proposed model outperforms state-of-the-art models in density estimation and representation learning.

Copied to clipboard

Kanzhi Cheng, Wenpo Song, Jiaxin Fan, Zheng Ma, Qiushi Sun, Fangzhi Xu, Chenyang Yan, Nuo Chen, Jianbing Zhang, Jiajun Chen

XToM: Exploring the Multilingual Theory of Mind for Large Language Models (2026.acl-long)

Challenge:	Image captioning has been a challenge for vision-language researchers for decades . current VLMs focus on tasks like visual question answering (YA) but image captioning is not as advanced as expected.
Approach:	They evaluate VLMs' performance on image captioning using human annotations . they find that some metrics show high caption-level agreement with humans .
Outcome:	The proposed model outperforms open-source models on image captioning . it achieves 93.4% correlation with human rankings at $4 per test .

Copied to clipboard

Chunkit Chan, Yauwai Yim, Hongchuan Zeng, Zhiying Zou, Xinyuan Cheng, Zhifan Sun, Zheye Deng, Kawai Chung, Yuzhuo Ao, Fan Yixiang, Cheng Jiayang, Ercong Nie, Ginny Wong, Helmut Schmid, Hinrich Schuetze, Simon See, Yangqiu Song

Can Large Language Models Grasp Legal Theories? Enhance Legal Reasoning with Insights from Multi-Agent Collaboration (2024.findings-emnlp)

Challenge:	Existing evaluations of ToM in LLMs are limited to English, neglecting the linguistic diversity that shapes human cognition.
Approach:	They propose a multilingual benchmark that evaluates ToM across five languages . they find that models excel in multilingual language understanding, but their ToM performance varies across languages.
Outcome:	The proposed benchmark evaluates LLMs across five languages and incorporates diverse task scenarios.

Copied to clipboard

Weikang Yuan, Junjie Cao, Zhuoren Jiang, Yangyang Kang, Jun Lin, Kaisong Song, Tianqianjin Lin, Pengwei Yan, Changlong Sun, Xiaozhong Liu

Geoparsing: Diagram Parsing for Plane and Solid Geometry with a Unified Formal Language (2026.findings-acl)

Challenge:	Existing studies have found that when LLMs are given criminal facts and legal rules, then asked whether cases constitute a certain charge, they struggle to understand legal theories and perform basic legal reasoning tasks.
Approach:	They propose a task to assess LLMs' understanding of legal theories and reasoning capabilities by using a novel framework: Multi-Agent framework for improving complex legal reasoning capability.
Outcome:	The proposed framework improves LLMs' understanding of legal theories and reasoning abilities in real-world scenarios.

Copied to clipboard

Peijie Wang, Ming-Liang Zhang, Jun Cao, Chao Deng, Dekang Ran, Pi Bu, Hongda Sun, Xuan Zhang, Yingyao Wang, Jun Song, Bo Zheng, Fei Yin, Cheng-Lin Liu

Reverse Chain: A Generic-Rule for LLMs to Master Multi-API Planning (2024.findings-naacl)

Challenge:	Recent advances in Multimodal Large Language Models (MLLMs) have demonstrated remarkable capabilities across various vision reasoning tasks.
Approach:	They propose a unified formal language that integrates plane and solid geometry, comprehensively covering geometric structures and semantic relations.
Outcome:	The proposed language achieves state-of-the-art parsing performance and significantly boosts MLLMs’ capabilities for downstream geometry reasoning tasks.

Copied to clipboard

Yinger Zhang, Hui Cai, Xierui Song, Yicheng Chen, Rui Sun, Jing Zheng

IPS: In-Prompt Process Supervision for Short Video Content Moderation (2026.acl-industry)

Challenge:	Existing approaches to enable large language models to implement function calling are limited in their tool-use capabilities.
Approach:	They propose a controllable, target-driven approach to empower LLMs to operate external APIs only via prompts.
Outcome:	The proposed approach limits LLMs to executing simple tasks, e.g., API Selection and Argument Completion.

Copied to clipboard

Mingchao Liu, Yu Sun, Ruixiao Sun, Xin Dong, Xiang Shen, Hongwei Wang, Hongyu Xiong, Yang Song

Act as you think: Reinforcing Consistent Reasoning in Medical Visual Question Answering (2026.acl-long)

Challenge:	Multimodal large language models (MLLMs) capture semantics of short video content but fail to account for policy-specific details.
Approach:	They propose a framework that integrates In-prompt Process Supervision into MLLMs . they propose sequential reasoning over ancillary questions during fine-tuning .
Outcome:	IPS outperforms baseline MLLMs on public and proprietary benchmarks . replacing human-annotated ancillary labels with MLML-generated ones results in performance degradation.

Copied to clipboard

Songtao Jiang, Yuan Wang, Ruizhe Chen, Yan Zhang, Ruilin Luo, Bohan Lei, Yeying Jin, Sibo Song, ZhiBo Yang, Jimeng Sun, Jian Wu, Zuozhu Liu

Robustness in NLP

Challenge:	Recent advances have improved the accuracy of medical visual question answering (Med-VQA) however, the high stakes nature of the medical domain has precipitated a shift towards interpretability and transparency of reasoning processes.
Approach:	They propose a reinforcement learning from verifiable rewards framework that rewards internal consistency and logical coherence.
Outcome:	The proposed framework rewards internal consistency and logical coherence, and is highly versatile, the authors show.

Cost-Optimal Grouped-Query Attention for Long-Context Modeling (2025.emnlp-main)

Copied to clipboard

Yingfa Chen, Yutong Wu, Chenyang Song, Zhen Leng Thai, Xingyu Shen, Xu Han, Zhiyuan Liu, Maosong Sun

Green & Sustainable NLP

Challenge:	Current GQA configurations overlook how context length influences inference cost .
Approach:	They propose a recipe for deriving cost-optimal GQA configurations that decouple the total head size from the hidden size and allow more flexible control over attention FLOPs.
Outcome:	The proposed configurations reduce memory usage and FLOPs by more than 50% compared to Llama-3's GQA, with no degradation in model capabilities.

RESPROMPT: Residual Connection Prompting Advances Multi-Step Reasoning in Large Language Models (2024.naacl-long)

Copied to clipboard

Song Jiang, Zahra Shakeri, Aaron Chan, Maziar Sanjabi, Hamed Firooz, Yinglong Xia, Bugra Akyildiz, Yizhou Sun, Jinchao Li, Qifan Wang, Asli Celikyilmaz

Multi-Turn Dialogue Generation in E-Commerce Platform with the Context of Historical Dialogue (2020.findings-emnlp)

Challenge:	Chain-of-thought (CoT) has impressively unlocked the reasoning potential of large language models (LLMs), but it falls short when tackling problems that require multiple reasoning steps.
Approach:	They propose a new prompting strategy that advances multi-step reasoning in LLMs by integrating necessary connections into prompts.
Outcome:	The proposed strategy improves multi-step reasoning accuracy and improves reasoning accuracy across math, sequential, and commonsense domains.

Copied to clipboard

WeiSheng Zhang, Kaisong Song, Yangyang Kang, Zhongqing Wang, Changlong Sun, Xiaozhong Liu, Shoushan Li, Min Zhang, Luo Si

HAHE: Hierarchical Attention for Hyper-Relational Knowledge Graphs in Global and Local Level (2023.acl-long)

Challenge:	Existing research on customer service dialogue generation generates generic responses from sellers . however, such cost prohibits small businesses, and multiturn dialogue generation is becoming more popular.
Approach:	They propose a novel and extensible dialogue generation method by leveraging sellers’ historical dialogue information to generate generic seller responses.
Outcome:	The proposed model can generate high-quality responses that cater to specific sellers’ characteristics and exhibit consistent superiority over baselines on a real-world multi-turn customer service dialogue dataset.

Copied to clipboard

Haoran Luo, Haihong E, Yuhao Yang, Yikai Guo, Mingzhi Sun, Tianyu Yao, Zichen Tang, Kaiyang Wan, Meina Song, Wei Lin

ResLoRA: Identity Residual Mapping in Low-Rank Adaption (2024.findings-acl)

Challenge:	Existing research on HKGs rarely models the graphical and sequential structure of HKG, limiting their representation.
Approach:	They propose a Hierarchical Attention model for HKG Embedding that includes global-level and local-level attention to model the graphical structure of HKGs.
Outcome:	The proposed model achieves state-of-the-art performance on HKG standard datasets and addresses the issue of HKG multi-position prediction for the first time.

Copied to clipboard

Shuhua Shi, Shaohan Huang, Minghui Song, Zhoujun Li, Zihan Zhang, Haizhen Huang, Furu Wei, Weiwei Deng, Feng Sun, Qi Zhang

Challenge:	Low-rank adaptation (LoRA) is one of the most popular parameter-efficient fine-tuning methods.
Approach:	They propose a low-rank adaptation method that adds residual paths during training and merges them together during inference to achieve better results.
Outcome:	The proposed method achieves 2.5x faster convergence speed and improves performance by 14.3% on NLG, NLU, and text-to-image tasks.

Scene Graph and Dependency Grammar Enhanced Remote Sensing Change Caption Network (SGD-RSCCN) (2025.coling-main)

Copied to clipboard

Qiaoli Sun, Yan Wang, Xiaoyu Song

STINMatch: Semi-Supervised Semantic-Topological Iteration Network for Financial Risk Detection via News Label Diffusion (2023.emnlp-main)

Challenge:	Remote sensing images are used for disaster assessment, urban planning and disaster response.
Approach:	They propose a Scene Graph and Dependency Grammar Enhanced Remote Sensing Change Caption Network to improve the accuracy and naturalness of extracting and describing change information from remote sensing images.
Outcome:	The proposed method improves the naturalness and accuracy of extracting and describing change information from remote sensing images.

Copied to clipboard

Xurui Li, Yue Qin, Rui Zhu, Tianqianjin Lin, Yongming Fan, Yangyang Kang, Kaisong Song, Fubang Zhao, Changlong Sun, Haixu Tang, Xiaozhong Liu

PDAMeta: Meta-Learning Framework with Progressive Data Augmentation for Few-Shot Text Classification (2024.lrec-main)

Challenge:	Commercial news provides rich semantics and timely information for automated financial risk detection.
Approach:	They propose a semi-supervised Semantic-Topological Iteration Network, STINMatch, along with a news-enterprise knowledge graph to endorse the risk detection enhancement.
Outcome:	The proposed model outperforms existing models in terms of generalization and semantics and annotation.

Copied to clipboard

Xurui Li, Kaisong Song, Tianqianjing Lin, Yangyang Kang, Fubang Zhao, Changlong Sun, Xiaozhong Liu

Low-Resource NLP

Challenge:	Existing text data augmentation methods can not ensure the diversity and quality of the generated data, which leads to sub-optimal performance.
Approach:	They propose a meta-learning framework with progressive data augmentation for few-shot text classification using prompt-based data augmented by attention-based methods.
Outcome:	The proposed framework outperforms state-of-the-art models and shows better robustness on four public few-shot text classification datasets.

MuKA: Multimodal Knowledge Augmented Visual Information-Seeking (2025.coling-main)

Copied to clipboard

Lianghao Deng, Yuchong Sun, Shizhe Chen, Ning Yang, Yunfeng Wang, Ruihua Song

Rethinking the Video Sampling and Reasoning Strategies for Temporal Sentence Grounding (2022.findings-emnlp)

Challenge:	Existing methods for visual information-seeking tasks rely on textual knowledge . existing methods can impair information retrieval and confuse MLLMs .
Approach:	They propose a framework which leverages a multimodal knowledge base to address these limitations.
Outcome:	The proposed framework outperforms state-of-the-art methods on the InfoSeek and E-VQA benchmarks.

Copied to clipboard

Jiahao Zhu, Daizong Liu, Pan Zhou, Xing Di, Yu Cheng, Song Yang, Wenzheng Xu, Zichuan Xu, Yao Wan, Lichao Sun, Zeyu Xiong

Train in Vain: Functionality-Preserving Poisoning to Prevent Unauthorized Use of Code Datasets (2026.findings-acl)

Challenge:	Existing methods for temporal sentence grounding ignore two crucial issues . 1) Boundary-bias: the video downsampling process may lose these two frames . 2) Reasoning-biases: such incorrect new boundary frames lead to the reasoning bias .
Approach:	They propose a siamese sampling mechanism to generate additional contextual frames . they use a reasoning strategy to learn the inter-relationship among these frames a .
Outcome:	Extensive experiments demonstrate the effectiveness of a new siamese sampling network on three challenging datasets.

Copied to clipboard

Yuan Xiao, Jiaming Wang, Yuchen Chen, Wei Song, Jun Sun, Shiqing Ma, Yanzhou Mu, Juan Zhai, Chunrong Fang, Jin Song Dong, Zhenyu Chen

Programming Languages in NLP

Challenge:	Existing methods for dataset poisoning require full-dataset poison, which breaks code compilability.
Approach:	They propose a functionality-preserving poisoning approach that injects short, compilable weak-use fragments into executed code paths.
Outcome:	The proposed method contaminates 10% of the dataset while maintaining 100% compilability and functional correctness.

Parrot: A Training Pipeline Enhances Both Program CoT and Natural Language CoT for Reasoning (2025.emnlp-main)

Copied to clipboard

Senjie Jin, Lu Chen, Zhiheng Xi, Yuhui Wang, Sirui Song, Yuhao Zhou, Xinbo Zhang, Peng Sun, Hong Lu, Tao Gui, Qi Zhang, Xuanjing Huang

A Role-Selected Sharing Network for Joint Machine-Human Chatting Handoff and Service Satisfaction Analysis (2021.emnlp-main)

Challenge:	Existing work focuses on enabling models to generate natural language chain-of-thought rationales or leverage executable and verifiable code, such as Python.
Approach:	They propose a novel training pipeline that integrates sequential P-CoT and N-Co T generation and a subtask hybrid training strategy to facilitate natural language transferability.
Outcome:	The proposed training pipeline improves both N-CoT and P-Co T performance over the RL baseline.

Copied to clipboard

Jiawei Liu, Kaisong Song, Yangyang Kang, Guoxiu He, Zhuoren Jiang, Changlong Sun, Wei Lu, Xiaozhong Liu

AceGPT, Localizing Large Language Models in Arabic (2024.naacl-long)

Challenge:	Recent efforts to predict chatbot failure hatches vital apprehensions due to complexity of human conversation.
Approach:	They propose a model that integrates dialogue satisfaction estimation and handoff prediction in one multi-task learning framework.
Outcome:	The proposed model integrates dialogue satisfaction estimation and handoff prediction in one multi-task learning framework.

Copied to clipboard

Huang Huang, Fei Yu, Jianqing Zhu, Xuening Sun, Hao Cheng, Song Dingjie, Zhihong Chen, Mosen Alharthi, Bang An, Juncai He, Ziche Liu, Junying Chen, Jianquan Li, Benyou Wang, Lian Zhang, Ruoyu Sun, Xiang Wan, Haizhou Li, Jinchao Xu

DPGA-TextSyn: Differentially Private Genetic Algorithm for Synthetic Text Generation (2025.findings-acl)

Challenge:	Significant concerns emerge when addressing cultural sensitivity and local values.
Approach:	They propose a localized Large Language Model (LLM) specifically for Arabic, a language imbued with unique cultural characteristics inadequately addressed by current mainstream models.
Outcome:	The proposed model sets the state-of-the-art standard for open Arabic LLMs across various benchmarks.

Copied to clipboard

Zhonghao Sun, Zhiliang Tian, Yiping Song, Yuyi Si, Juhua Zhang, Minlie Huang, Kai Lu, Zeyu Xiong, Xinwang Liu, Dongsheng Li

Ethical NLP

Challenge:	Existing methods to fine-tune large language models pose privacy risks . researchers have synthesized data with strong generation capabilities closed-source LLMs to alleviate this problem .
Approach:	They propose to combine general LLMs with genetic algorithm to produce relevant and diverse synthetic text under differential privacy constraints.
Outcome:	The proposed method significantly improves the performance of the model in downstream tasks while ensuring privacy.

Tracking Satisfaction States for Customer Satisfaction Prediction in E-commerce Service Chatbots (2022.coling-1)

Copied to clipboard

Yang Sun, Liangqing Wu, Shuangyong Song, Xiaoguang Yu, Xiaodong He, Guohong Fu

AutoSDT: Scaling Data-Driven Discovery Tasks Toward Open Co-Scientists (2025.emnlp-main)

Challenge:	Existing models for customer satisfaction prediction (CSP) focus on analyzing subjective customer satisfaction in conversational service, but they are hard to represent the important dynamic satisfaction states throughout the customer journey.
Approach:	They propose a model to track customer satisfaction in chatbots using a dialogue-level classification module to represent the dynamic satisfaction states at each turn.
Outcome:	The proposed model outperforms baselines and shows that it significantly outperformed multiple baselines.

Copied to clipboard

Yifei Li, Hanane Nour Moussa, Ziru Chen, Shijie Chen, Botao Yu, Mingyi Xue, Benjamin Burns, Tzu-Yao Chiu, Vishal Dey, Zitong Lu, Chen Wei, Qianheng Zhang, Tianyu Zhang, Song Gao, Xuhui Huang, Xia Ning, Nesreen K. Ahmed, Ali Payani, Huan Sun

A Length-Extrapolatable Transformer (2023.acl-long)

Challenge:	AutoSDT-5K is the only automatically collected and the largest open dataset for data-driven scientific discovery.
Approach:	They propose an automatic pipeline that collects high-quality coding tasks in real-world data-driven discovery workflows.
Outcome:	The proposed pipeline synthesizes accurate tasks and tasks from a dataset of 5,404 tasks covering four scientific disciplines and 756 Python packages.

Copied to clipboard

Yutao Sun, Li Dong, Barun Patra, Shuming Ma, Shaohan Huang, Alon Benhaim, Vishrav Chaudhary, Xia Song, Furu Wei

Context-Value-Action Architecture for Value-Driven Large Language Model Agents (2026.findings-acl)

Challenge:	Existing Transformers can only deal with the in-distribution size of inputs.
Approach:	They propose a relative position embedding to explicitly maximize attention resolution . they also use blockwise causal attention during inference for better resolution a .
Outcome:	The proposed model achieves strong performance in interpolation and extrapolation settings.

Copied to clipboard

TianZe Zhang, Sirui Sun, Yuhang Xie, Xin Zhang, Zhiqiang Wu, Guojie Song

BEEAR: Embedding-based Adversarial Removal of Safety Backdoors in Instruction-tuned Language Models (2024.emnlp-main)

Challenge:	Existing LLMs exhibit behavioral rigidity, a flaw often masked by the self-referential bias of current "LLM-as-a-judge" evaluations.
Approach:	They propose a Context-Value-Action architecture that decouples action generation from cognitive reasoning via a Value Verifier trained on authentic human data to explicitly model dynamic value activation.
Outcome:	The proposed architecture significantly outperforms baseline models on 1.1 million real-world interaction traces on CVABench.

Copied to clipboard

Yi Zeng, Weiyu Sun, Tran Huynh, Dawn Song, Bo Li, Ruoxi Jia

FUSE: Measure-Theoretic Compact Fuzzy Set Representation for Taxonomy Expansion (2024.findings-acl)

Challenge:	Safety backdoors in large language models can be triggered while evading detection during normal interactions.
Approach:	They propose a bi-level optimization method that uses a key insight: backdoor triggers induce a uniform drift in the model’s embedding space . inner level identifies universal perturbations to the decoder’s embedded spaces that steer the model towards defender-defined unwanted behaviors; outer level fine-tunes the model to reinforce safe behaviors against these perturbations.
Outcome:	The proposed mitigation method reduces the success rate of safety backdoor attacks from over 95% to 1% for general harmful behaviors and from 47% to 0% for Sleeper Agents, without compromising the model’s usefulness.

Copied to clipboard

Fred Xu, Song Jiang, Zijie Huang, Xiao Luo, Shichang Zhang, Yuanzhou Chen, Yizhou Sun

Multi-Modal Multi-Granularity Tokenizer for Chu Bamboo Slips (2025.coling-main)

Challenge:	Existing work models taxonomy concepts as vectors or geometric objects, but fuzzy sets are efficient for concept modeling.
Approach:	They propose a set representation learning task based on fuzzy set approximation . they demonstrate remarkable improvements in taxonomy expansion using FUSE .
Outcome:	The proposed framework improves taxonomy expansion performance by 23% over baselines.

Copied to clipboard

Yingfa Chen, Chenlong Hu, Cong Feng, Chenyang Song, Shi Yu, Xu Han, Zhiyuan Liu, Maosong Sun

Syntactic Text Processing

Challenge:	Using a multi-modal multi-granularity tokenizer, we analyze ancient Chinese scripts . a large proportion of the characters in ancient Chinese are rare or undeciphered .
Approach:	They propose a multi-modal multi-granularity tokenizer specifically designed for ancient Chinese scripts.
Outcome:	The proposed tokenizer improves on the part-of-speech tagging task on the Chu bamboo slip script.

Hallucination Detection in Long-Form Text Generated by LLMs: A Benchmark and a Hyper-Relational Knowledge Graph Approach (2026.findings-acl)

Copied to clipboard

Zituo Li, Guangzhou Chen, Jianbin Sun, Qi Song, Yang Kewei

CritiQ: Mining Data Quality Criteria from Human Preferences (2025.acl-long)

Challenge:	Existing methods for hallucination detection are coarse-grained and lack long-range consistency checks.
Approach:	They propose a benchmark for long-form hallucination detection that incorporates diverse entity types and intricate factual dependencies spanning extended contexts.
Outcome:	The proposed framework outperforms baselines and robustly integrates fact-centric hyper-relational knowledge graphs.

Copied to clipboard

Honglin Guo, Kai Lv, Qipeng Guo, Tianyi Liang, Zhiheng Xi, Demin Song, Qiuyinzhe Zhang, Yu Sun, Kai Chen, Xipeng Qiu, Tao Gui

SimpleDeepSearcher: Deep Information Seeking via Web-Powered Reasoning Trajectory Synthesis (2025.findings-emnlp)

Challenge:	Existing methods to train language models rely on manual design, perplexity, or careful prompt engineering.
Approach:	They propose a method that automatically mines criteria from human preferences for data quality with only 30 human-annotated pairs and performs efficient data selection.
Outcome:	The proposed method improves on human-annotated test sets and shows high accuracy on code, math, and logic domains.

Copied to clipboard

Shuang Sun, Huatong Song, Yuhao Wang, Ruiyang Ren, Jinhao Jiang, Junjie Zhang, Fei Bai, Jia Deng, Xin Zhao, Zheng Liu, Lei Fang, Zhongyuan Wang, Ji-Rong Wen

Decoding at the Speed of Thought: Harnessing Parallel Decoding of Lexical Units for LLMs (2024.lrec-main)

Challenge:	Existing approaches to deep search training lack high-quality training trajectories, prohibitive computational costs and lack of high-fidelity training data.
Approach:	They propose a framework that synthesizes high-quality training data by simulating real user interactions in live web search environments.
Outcome:	The proposed framework synthesizes high-quality training data by simulating user interactions in live web search environments.

Copied to clipboard

Chenxi Sun, Hongzhi Zhang, Zijia Lin, Jingyuan Zhang, Fuzheng Zhang, Zhongyuan Wang, Bin Chen, Chengru Song, Di Zhang, Kun Gai, Deyi Xiong

Divide-Then-Align: Honest Alignment based on the Knowledge Boundary of RAG (2025.acl-long)

Challenge:	Large language models have demonstrated exceptional capability in natural language understanding and generation, but their generation speed is limited by the inherently sequential nature of their decoding process.
Approach:	They propose a method that accelerates decoding process without sacrificing quality . they propose lexical unit decoding, which can be integrated with other methods .
Outcome:	The proposed method significantly reduces decoding time while maintaining quality while maintaining output quality.

Copied to clipboard

Xin Sun, Jianan Xie, Zhongqi Chen, Qiang Liu, Shu Wu, Yuehe Chen, Bowen Song, Zilei Wang, Weiqiang Wang, Liang Wang

Parrot: Enhancing Multi-Turn Instruction Following for Large Language Models (2024.acl-long)

Challenge:	Large language models (LLMs) augmented with retrieval systems have significantly advanced natural language processing tasks by integrating external knowledge sources.
Approach:	They propose a method that conditions large language models to generate answers even in the absence of reliable knowledge.
Outcome:	The proposed approach balances accuracy with appropriate abstention, enhancing the reliability and trustworthiness of retrieval-augmented systems.

Copied to clipboard

Yuchong Sun, Che Liu, Kun Zhou, Jinwen Huang, Ruihua Song, Xin Zhao, Fuzheng Zhang, Di Zhang, Kun Gai

MIRTH: Mutual-Information Reasoning with Temporal Hubs for Vision-Language-Action Agents (2026.acl-long)

Challenge:	Existing studies overlook the multi-turn instruction following ability of large language models (LLMs) Extensive experiments show that Parrot improves current LLMs by up to 7.2% in multi- turn instruction following.
Approach:	They propose a method for collecting multi-turn instructions that feature human-like queries, such as anaphora and ellipsis, and a context-aware preference optimization strategy to further enhance LLMs for complex queries.
Outcome:	The proposed method improves existing LLMs by up to 7.2% in multi-turn instruction following.

Copied to clipboard

Hao Sun, Yu Song, Teng Shiyu, Ziwei Niu, Yen-wei Chen

ProSparse: Introducing and Enhancing Intrinsic Activation Sparsity within Large Language Models (2025.coling-main)

Challenge:	a recent study shows that VLA models suffer from temporal myopia that discards historical dynamics and reasoning gaps between high-level instructions and low-level motor commands.
Approach:	They propose a framework to address temporal myopia and autoregressive scalar decoding in VLAs . they propose two memory hubs that compress long-term scene evolution and short-term motion trends .
Outcome:	The proposed framework achieves state-of-the-art performance and exhibiting emergent error recovery capabilities.

Copied to clipboard

Chenyang Song, Xu Han, Zhengyan Zhang, Shengding Hu, Xiyu Shi, Kuai Li, Chen Chen, Zhiyuan Liu, Guangli Li, Tao Yang, Maosong Sun

Multimodal Machine Translation with Text-Image In-depth Questioning (2025.findings-acl)

Challenge:	Activation sparsity is a promising paradigm for accelerating model inference . few large language models achieve high activation spar and comparable performance .
Approach:	They propose a method to achieve activation sparsity and acceleration in large language models . they introduce ReLU activation and adopt progressive sparse regularization .
Outcome:	The proposed method achieves high activation sparsity and comparable model performance.

Copied to clipboard

Yue Gao, Jing Zhao, Shiliang Sun, Xiaosong Qiao, Tengfei Song, Hao Yang