Papers by Ke Li

M3Seg: A Maximum-Minimum Mutual Information Paradigm for Unsupervised Topic Segmentation in ASR Transcripts (2023.emnlp-main)

Challenge:	Large language models are trained on static corpora but deployed in a dynamic world . a foundational tension remains between time and the ability to understand it .
Approach:	They formalize temporal queries in an information-theoretic framework based on parametric reachability of temporal premises and answers.
Outcome:	The proposed framework formalizes temporal queries in an information-theoretic framework based on parametric reachability of temporal premises and answers . the framework induces four temporal information regimes corresponding to internal reasoning, answer recency, premise anchoring, and genuine world indeterminacy .

Copied to clipboard

Ke Wang, Xiutian Zhao, Yanghui Li, Wei Peng

Text Generation

Challenge:	Topic segmentation aims to split automatic speech recognition transcriptions into segments that are bounded by thematic meanings.
Approach:	They propose a Maximum-Minimum Mutual information paradigm for linear topic segmentation without using any parallel data.
Outcome:	The proposed paradigm outperforms the state-of-the-art methods by a significant margin.

MathGenie: Generating Synthetic Data with Question Back-translation for Enhancing Mathematical Reasoning of LLMs (2024.acl-long)

Copied to clipboard

Zimu Lu, Aojun Zhou, Houxing Ren, Ke Wang, Weikang Shi, Junting Pan, Mingjie Zhan, Hongsheng Li

DOC-RAG: ASR Language Model Personalization with Domain-Distributed Co-occurrence Retrieval Augmentation (2024.lrec-main)

Challenge:	Existing models have demonstrated outstanding capabilities in mathematical reasoning, but there is a performance gap between open-source models and closed-source ones.
Approach:	They propose a method for generating diverse and reliable math problems by leveraging the ground-truth solutions of the seed data.
Outcome:	The proposed model outperforms open-source models across five representative mathematical reasoning datasets.

Copied to clipboard

Puneet Mathur, Zhe Liu, Ke Li, Yingyi Ma, Gil Karen, Zeeshan Ahmed, Dinesh Manocha, Xuedong Zhang

Efficient Inference for Large Vision-Language Models: Bottlenecks, Techniques, and Prospects (2026.findings-acl)

Challenge:	Extensive experiments on three user-specific speech-to-text tasks show that DOC-RAG significantly outperforms strong baselines with an 8-15% improvement in terms of perplexity and a 4-7% reduction in terms in terms . of Word Error Rates.
Approach:	They propose a domain-distributed co-occurrence augmentation approach to improve automatic speech recognition of rare word patterns in unseen domains by using n-gram co-existence distributions.
Outcome:	Experiments on three user-specific speech-to-text tasks show that DOC-RAG outperforms baselines with an 8-15% improvement in terms of perplexity and a 4-7% reduction in terms in terms . of Word Error Rates.

Copied to clipboard

Jun Zhang, Yicheng Ji, Feiyang Ren, Yihang Li, Bowen Zeng, Zonghao Chen, Ke Chen, Lidan Shou, Gang Chen, Huan Li

P-INT: A Path-based Interaction Model for Few-shot Knowledge Graph Completion (2021.findings-emnlp)

Challenge:	Large Vision-Language Models are hindered by a systemic efficiency barrier known as visual token dominance.
Approach:	They propose a systematic taxonomy of efficiency techniques structured around the inference lifecycle . they examine visual encoding, prefilling, and decoding to understand bottlenecks .
Outcome:	The proposed techniques reveal how upstream decisions dictate downstream bottlenecks . the proposed techniques include hybrid compression and modality-aware decoding .

Copied to clipboard

Jingwen Xu, Jing Zhang, Xirui Ke, Yuxiao Dong, Hong Chen, Cuiping Li, Yongbin Liu

CodeArena: Evaluating and Aligning CodeLLMs on Human Preference (2025.emnlp-main)

Challenge:	Existing methods to encode and match entity pairs have only a few observed reference entity pairs.
Approach:	They propose a model that infers and leverages paths that can expressively encode the relation of two entities.
Outcome:	The proposed model outperforms the state-of-the-art models by 11.2– 14.2% in terms of Hits@1.

Copied to clipboard

Jian Yang, Jiaxi Yang, Wei Zhang, Jin Ke, Yibo Miao, Lei Zhang, Liqun Yang, Zeyu Cui, Yichang Zhang, Zhoujun Li, Binyuan Hui, Junyang Lin

AgentBank: Towards Generalized LLM Agents via Fine-Tuning on 50000+ Interaction Trajectories (2024.findings-emnlp)

Challenge:	Code large language models (codeLLMs) focus on synthesizing the correct code snippet, ignoring the alignment with human preferences.
Approach:	They propose a benchmark code-based on 40 categories and 44 programming languages to emulate real-world coding tasks.
Outcome:	The proposed benchmarks show that open-source code LLMs perform better than open-sourced ones.

Copied to clipboard

Yifan Song, Weimin Xiong, Xiutian Zhao, Dawei Zhu, Wenhao Wu, Ke Wang, Cheng Li, Wei Peng, Sujian Li

Enhancing Persona Consistency for LLMs’ Role-Playing using Persona-Aware Contrastive Learning (2025.findings-acl)

Challenge:	Existing studies focus on specialized agents designed for particular tasks.
Approach:	They propose to scale annotated interaction trajectories and fine-tune LLMs on AgentBank to get a series of agent models, Samoyed.
Outcome:	The proposed model can scale to get generalized agent capabilities.

Copied to clipboard

Ke Ji, Yixin Lian, Linxu Li, Jingsheng Gao, Weiyuan Li, Bin Dai

One QuantLLM for ALL: Fine-tuning Quantized LLMs Once for Efficient Deployments (2025.acl-long)

Challenge:	Existing methods for analyzing and analyzing large language models (LLMs) lack of emotion and fine-grained role awareness limits the model’s ability to provide personalized and diverse interactions further.
Approach:	They propose an annotation-free framework to align LLMs’ behavior during role-playing, enhancing the model’s role consistency.
Outcome:	The proposed framework outperforms vanilla LLMs under automatic evaluation methods and human expert evaluation.

Copied to clipboard

Ke Yi, Yuhui Xu, Heng Chang, Yuan Meng, Tong Zhang, Jia Li

Few-Shot Multimodal Named Entity Recognition Based on Mutlimodal Causal Intervention Graph (2024.lrec-main)

Challenge:	Quantization has shown promise for Large Language Models, but current methods require lengthy training to alleviate quantization loss.
Approach:	They propose to decouple weights and incorporate Low-Rank adapters to reduce weight sharing . they validate the approach on LLaMA2 families and Mistral on downstream evaluation .
Outcome:	The proposed approach shows high performance while reducing deployment time faced with multiple scenarios.

Copied to clipboard

Feihong Lu, Xiaocui Yang, Qian Li, Qingyun Sun, Ke Jiang, Cheng Ji, Jianxin Li

Draft & Verify: Lossless Large Language Model Acceleration via Self-Speculative Decoding (2024.acl-long)

Challenge:	Existing methods for multimodal named entity recognition are limited due to limited resources.
Approach:	They propose a Few-shot Multimodal Named Entity Recognition task to address these relation types by constructing a multimodal graph and a new multimodal causal intervention strategy.
Outcome:	The proposed model improves on two multimodal named entity recognition datasets.

Copied to clipboard

Jun Zhang, Jue Wang, Huan Li, Lidan Shou, Ke Chen, Gang Chen, Sharad Mehrotra

TS-Align: A Teacher-Student Collaborative Framework for Scalable Iterative Finetuning of Large Language Models (2024.findings-emnlp)

Challenge:	Existing methods for accelerating Large Language Models have been criticized for their inference costs and inefficient decoding.
Approach:	They propose a self-speculative decoding approach for accelerating Large Language Models without an auxiliary model.
Outcome:	The proposed method achieves a speedup of up to 1.99 with no additional neural network training and no extra memory footprint.

Copied to clipboard

Chen Zhang, Chengguang Tang, Dading Chong, Ke Shi, Guohua Tang, Feng Jiang, Haizhou Li

On the Consistency of Commonsense in Large Language Models (2025.findings-acl)

Challenge:	Mainstream approaches to aligning large language models heavily rely on human preference data.
Approach:	They propose a framework that fine-tunes a policy model using pairwise feedback data automatically mined from its outputs.
Outcome:	The proposed framework outperforms the base model with an average win rate of 69.7% across seven conversational or instruction-following datasets.

Copied to clipboard

Guozheng Li, Peng Wang, Wenjun Ke, Zijie Xu, Jiajun Liu, Ziyu Shang

Acquisition and Application of Novel Knowledge in Large Language Models (2025.acl-long)

Challenge:	Existing evaluations of commonsense for large language models focus on downstream knowledge tasks, failing to probe whether LLMs truly understand and utilize knowledge or merely memorize it.
Approach:	They propose to automatically construct a large benchmark named CoCo which measures LLMs’ knowledge memorization, comprehension, and application and examines the consistency between these tasks.
Outcome:	The proposed benchmark systematically assesses LLMs’ knowledge memorization, comprehension, and application and examines the consistency between these tasks.

Copied to clipboard

Ziyu Shang, Jianghan Liu, Zhizhao Luo, Peng Wang, Wenjun Ke, Jiajun Liu, Zijie Xu, Guozheng Li

MP-RNA: Unleashing Multi-species RNA Foundation Model via Calibrated Secondary Structure Prediction (2024.findings-emnlp)

Challenge:	Existing methods for constructing new datasets rely on timestamps or simple templates that do not accurately reflect the real world.
Approach:	They propose a knowledge dataset construction approach that simulates biological evolution using knowledge graphs to generate synthetic entities with diverse attributes.
Outcome:	The proposed framework outperforms knowledge augmentation methods by 3.3%-38%.

Copied to clipboard

Knowledge-augmented Self-training of A Question Rewriter for Conversational Knowledge Base Question Answering (2022.findings-emnlp)

Challenge:	Experimental evaluations demonstrate that our RNA FM consistently outperforms existing RNA .
Approach:	They propose to use filtered high-fidelity structure annotations to enhance the modeling ability of FMs in single nucleotide resolution tasks.
Outcome:	The proposed model outperforms existing RNA FMs on four genomic benchmarks and achieves top-tier results on DNA genomic benchmark.

Copied to clipboard

Xirui Ke, Jing Zhang, Xin Lv, Yiqi Xu, Shulin Cao, Cuiping Li, Hong Chen, Juanzi Li

MathCoder-VL: Bridging Vision and Code for Enhanced Multimodal Mathematical Reasoning (2025.findings-acl)

Challenge:	Recent rise of conversational applications has promoted the development of conversation KBQA (ConvKBQA).
Approach:	They propose a framework to produce a full-fledged rewritten question based on conversation history and then reason the answer by existing single-turn KBQA models.
Outcome:	The proposed framework produces a full-fledged rewritten question based on the conversation history and reasoned the answer by existing single-turn KBQA models.

Copied to clipboard

Ke Wang, Junting Pan, Linda Wei, Aojun Zhou, Weikang Shi, Zimu Lu, Han Xiao, Yunqiao Yang, Houxing Ren, Mingjie Zhan, Hongsheng Li

Revisiting Large Language Models as Zero-shot Relation Extractors (2023.findings-emnlp)

Challenge:	Large Language Models (LMMs) struggle with simple tasks such as geometry, e.g., arithmetic, and reasoning.
Approach:	They propose to leverage code as supervision for cross-modal alignment . they propose to use FigCodifier and ImgCode-8.6M to synthesize novel mathematical figures .
Outcome:	The proposed model surpasses GPT-4o and Claude 3.5 Sonnet in the geometry problem-solving subset of MathVista, achieving improvements of 8.9% and 9.2%.

Copied to clipboard

Guozheng Li, Peng Wang, Wenjun Ke

ShieldLM: Empowering LLMs as Aligned, Customizable and Explainable Safety Detectors (2024.findings-emnlp)

Challenge:	Recent studies show that large language models (LLMs) transfer well to new tasks out-of-the-box . relationship extraction (RE) involves a certain degree of labeled or unlabeled data even under zero-shot setting.
Approach:	They propose a simple prompt recursively using LLMs to transform RE inputs to QA format . they propose qq prompting and qt prompting to improve their results .
Outcome:	The proposed method improves on different model sizes, benchmarks and settings.

Copied to clipboard

Zhexin Zhang, Yida Lu, Jingyuan Ma, Di Zhang, Rui Li, Pei Ke, Hao Sun, Lei Sha, Zhifang Sui, Hongning Wang, Minlie Huang

Responsible & Trustworthy NLP

Challenge:	Existing tools for detecting safety issues in LLMs are expensive and inefficient.
Approach:	They propose an LLM-based safety detector which annotates the safety of queries and provides explanations for its decisions.
Outcome:	The proposed detector outperforms baselines on four sets of query-response pairs and is effective as a safety evaluator for advanced LLMs.

xDial-Eval: A Multilingual Open-Domain Dialogue Evaluation Benchmark (2023.findings-emnlp)

Copied to clipboard

Chen Zhang, Luis D’Haro, Chengguang Tang, Ke Shi, Guohua Tang, Haizhou Li

HILL: Hierarchy-aware Information Lossless Contrastive Learning for Hierarchical Text Classification (2024.naacl-long)

Challenge:	Currently, human evaluation is the most reliable way to holistically judge the quality of the dialogue.
Approach:	They propose to use English dialogue evaluation metrics to generalize them to other languages.
Outcome:	The proposed metrics outperform OpenAI’s ChatGPT in terms of average Pearson correlations over all datasets and languages.

Copied to clipboard

He Zhu, Junran Wu, Ruomei Liu, Yue Hou, Ze Yuan, Shangzhe Li, Yicheng Pan, Ke Xu

Take the essence and discard the dross: A Rethinking on Data Selection for Fine-Tuning Large Language Models (2025.naacl-long)

Challenge:	Existing self-supervised methods in natural language processing rely on augmentation rules to generate contrastive samples.
Approach:	They propose a hierarchy-aware information lossless contrastive learning scheme that uses syntactic information reserved in the input sample and fused during the learning process.
Outcome:	The proposed learning scheme is superior to existing methods in hierarchical text classification . the proposed learning system is based on a structure encoder and a text encoder .

Copied to clipboard

Ziche Liu, Rui Ke, Yajiao Liu, Feng Jiang, Haizhou Li

SciAssess: Benchmarking LLM Proficiency in Scientific Literature Analysis (2025.findings-naacl)

Challenge:	Existing studies focus on data selection but lack a clear, unified framework . variability in experimental settings complicates systematic comparisons .
Approach:	They propose a three-stage scheme to standardize data selection for fine-tuning large language models . they propose unified comparison approach that incorporates ratio-based efficiency and ranking-based feasibility metrics to address inconsistencies across experiments.
Outcome:	The proposed scheme outperforms existing methods in a dozen key studies and identifies key challenges.

Copied to clipboard

Hengxing Cai, Xiaochen Cai, Junhan Chang, Sihang Li, Lin Yao, Wang Changxin, Zhifeng Gao, Hongshuai Wang, Li Yongge, Mujie Lin, Shuwen Yang, Jiankun Wang, Mingjun Xu, Jin Huang, Xi Fang, Jiaxi Zhuang, Yuqi Yin, Yaqi Li, Changhong Chen, Zheng Cheng, Zifeng Zhao, Linfeng Zhang, Guolin Ke

Challenge:	Existing benchmarks fail to adequately evaluate the proficiency of Large Language Models (LLMs) Existing standards do not cover the skills needed to evaluate LLMs in scientific literature analysis.
Approach:	They propose a benchmark to evaluate the proficiency of large language models in scientific literature analysis.
Outcome:	SciAssess evaluates 11 LLMs on multiple tasks across scientific fields.

QiMeng-PRepair: Precise Code Repair via Edit-Aware Reward Optimization (2026.acl-long)

Copied to clipboard

Changxin Ke, Rui Zhang, Jiaming Guo, Yuanbo Wen, Li Ding, Shuo Wang, Xuyuan Zhu, Xiong Peng, Di Huang, Zidong Du, Xing Hu, Qi Guo, Yunji Chen

Mixture-of-Minds: Multi-Agent Reinforcement Learning for Table Understanding (2026.acl-long)

Challenge:	Existing approaches to program repair are based on correctness alone.
Approach:	They propose a framework that mitigates over-editing and improves repair accuracy by generating buggy programs and re-edits.
Outcome:	The proposed framework improves repair precision by 31.4% under fix1@1, a metric that considers repair correctness and extent, and significantly increases decoding throughput when combined with speculative editing.

Copied to clipboard

Yuhang Zhou, Mingrui Zhang, Ke Li, Mingyi Wang, Qiao Liu, Qifei Wang, Jiayi Liu, Fei Liu, Serena Li, Weiwei LI, Mingze Gao, Abhishek Kumar, Xiangjun Fan, Zhuokai Zhao, Lizhu Zhang

Gated Mechanism Enhanced Multi-Task Learning for Dialog Routing (2022.coling-1)

Challenge:	Large language models (LLMs) have shown promise on understanding and reasoning over tables, but current approaches remain limited.
Approach:	They propose a multi-agent framework that decomposes table reasoning into three specialized roles: planning, coding, and answering.
Outcome:	The proposed framework decomposes table reasoning into three specialized roles: planning, coding, and answering.

Copied to clipboard

Ziming Huang, Zhuoxuan Jiang, Ke Wang, Juntao Li, Shanshan Feng, Xian-Ling Mao

SeqXGPT: Sentence-Level AI-Generated Text Detection (2023.emnlp-main)

Challenge:	Existing methods for dialog routing are mostly heuristic and cannot achieve high-quality performance.
Approach:	They propose a multi-task learning framework with a dialog encoder and two tailored gated mechanism modules to solve this problem.
Outcome:	The proposed model can play the role of hierarchical information filtering and is non-invasive to existing dialog systems.

Copied to clipboard

Pengyu Wang, Linyang Li, Ke Ren, Botian Jiang, Dong Zhang, Xipeng Qiu

Transfer-Aware Data Selection for Domain Adaptation in Text Retrieval (2025.findings-emnlp)

Challenge:	Existing methods for sentence-level AIGT detection are weak . large language models (LLMs) can generate human-like content .
Approach:	They propose a sentence-level AIGT detection challenge using LLMs as log probability lists . they propose 'check' GPT' method that uses log probability list features to detect AIGT .
Outcome:	The proposed method surpasses baseline methods in sentence- and document-level detection challenges.

Copied to clipboard

Linzhu Yu, Huan Li, Ke Chen, Lidan Shou

CTRLEval: An Unsupervised Reference-Free Metric for Evaluating Controlled Text Generation (2022.acl-long)

Challenge:	Existing methods to improve domain adaptation do not guarantee improved adaptability, but may negatively impact model performance.
Approach:	They propose a framework that can effectively improve model adaptability by selecting beneficial data without evaluating all source data.
Outcome:	The proposed framework improves model adaptability by selecting beneficial data without evaluating all source data.

Copied to clipboard

Pei Ke, Hao Zhou, Yankai Lin, Peng Li, Jie Zhou, Xiaoyan Zhu, Minlie Huang

Watermarking LLMs with Weight Quantization (2023.findings-emnlp)

Challenge:	Existing reference-free metrics have obvious limitations for evaluating controlled text generation models.
Approach:	They propose an unsupervised reference-free metric which evaluates controlled text generation from different aspects by formulating each aspect into multiple text infilling tasks.
Outcome:	The proposed metric has higher correlations with human judgments while obtaining better generalization of evaluating generated texts from different models and with different qualities.

Copied to clipboard

Linyang Li, Botian Jiang, Pengyu Wang, Ke Ren, Hang Yan, Xipeng Qiu

RACQC: Advanced Retrieval-Augmented Generation for Chinese Query Correction (2025.findings-emnlp)

Challenge:	Large language models are being deployed at an astonishing speed, exposing users to high risks.
Approach:	They propose a method that plants watermarks in quantization process of large language models without pre-defined triggers during inference.
Outcome:	The proposed method protects model weights without pre-defined triggers . it works when the model is used in the fp32 mode and remains hidden when the models are quantized to int8 .

Copied to clipboard

Jinbo Su, Lingzhe Gao, Wei Li, Shihao Liu, Haojie Lei, Xinyi Wang, Yuanzhao Guo, Ke Wang, Daiting Shi, Dawei Yin

Tree of Agents: Improving Long-Context Capabilities of Large Language Models through Multi-Perspective Reasoning (2025.findings-emnlp)

Challenge:	Large language models (LLMs) exhibit remarkable capabilities across many tasks, but face critical challenges in the CSC scenario: (1) poor generalization to rare entities in open-domain searches; and (2) failure to adapt to temporal entity variations due to static parameters, resulting in serious over-correction issues.
Approach:	They propose a Chinese Spelling Check system with RAG and multi-task learning that integrates dynamic knowledge retrieval and entity-centric RAG to address rare entities.
Outcome:	The proposed system outperforms existing baselines in the CSC task and achieves a maximum improvement of +9.92% on the search scenario benchmark and +3.2% on the general-domain dataset.

Copied to clipboard

Song Yu, Xiaofei Xu, Ke Deng, Li Li, Lin Tian

Diversifying Question Generation over Knowledge Base via External Natural Questions (2024.lrec-main)

Challenge:	Large language models face persistent challenges when handling long-context tasks . existing methods that reduce input have the risk of discarding key information .
Approach:	To address this issue, we propose a multi-agent reasoning framework called Tree of Agents . the framework segments input into chunks processed by independent agents .
Outcome:	The proposed model outperforms baseline models on long-context tasks.

Copied to clipboard

Shasha Guo, Jing Zhang, Xirui Ke, Cuiping Li, Hong Chen

Text Generation

Challenge:	Existing methods on knowledge base question generation focus on refining the quality of a single generated question.
Approach:	They propose a new diversity evaluation metric which measures the diversity among top-k generated questions for each instance while ensuring their relevance to the ground truth.
Outcome:	The proposed model outperforms pre-trained language model baselines and text-davinci-003 in diversity while achieving comparable performance with ChatGPT.

From Implicit Graph Encoding to Explicit Evidence: A Training-Free LLM Framework for Temporal Knowledge Graph Reasoning (2026.findings-acl)

Copied to clipboard

Guo Tang, Ke Cheng, Huiming Fan, Heng Chang, Wenxiang Zheng, Xianhao Ou, Junjia Xiang, Ming Liu, Yujun Zhou, Li Lanyu, Bing Qin

Sub-network Discovery and Soft-masking for Continual Learning of Mixed Tasks (2023.findings-emnlp)

Challenge:	Existing Large Language Models (LLMs) struggle with implicit modality alignment and suboptimal graph linearization.
Approach:	They propose a training-free, test-time adaptive framework that reframes TKG prediction as explicit evidence-driven reasoning.
Outcome:	ExE-LLM outperforms fully trained graph neural networks on four benchmarks . it achieves SOTA performance in inductive settings, significantly outperforming fully trained neural networks .

Copied to clipboard

Zixuan Ke, Bing Liu, Wenhan Xiong, Asli Celikyilmaz, Haoran Li

Benchmarking Vision-Language Models on Chinese Ancient Documents: From OCR to Knowledge Reasoning (2026.findings-acl)

Challenge:	Existing methods for Continual Learning (CL) have limited KT and catastrophic forgetting . a new method overcomes CF by isolating the knowledge of each task .
Approach:	They propose a method to overcome catastrophic forgetting and encourage knowledge transfer . they propose to discover a sub-network for each task and a soft-masking mechanism to preserve the previous knowledge.
Outcome:	The proposed method outperforms baselines in classification, generation, information extraction and their mixture.

Copied to clipboard

Haiyang Yu, Yuchuan Wu, Fan Shi, Jinghui Lu, Ke Niu, Xiaodong Ge, Minghan Zhuo, Jingqun Tang, Bin Li

McBE: A Multi-task Chinese Bias Evaluation Benchmark for Large Language Models (2025.findings-acl)

Challenge:	Existing document benchmarks focus on English printed texts or simplified Chinese . current vision-language models struggle with visual complexity and poor adaptability .
Approach:	They propose a benchmark to evaluate Chinese ancient documents' visual/linguistic complexity . ancient documents are valuable cultural heritage, but they face challenges in digitization and understanding .
Outcome:	the first benchmark for Chinese ancient documents evaluates VLMs from OCR to knowledge reasoning . ancient documents carry thousands of years of Chinese history and culture . traditional methods only scan images, while current models struggle with visual complexity .

Copied to clipboard

Tian Lan, Xiangdong Su, Xu Liu, Ruirui Wang, Ke Chang, Jiang Li, Guanglai Gao

Responsible & Trustworthy NLP

Challenge:	Existing datasets on bias evaluation for large language models focus on English and North American culture and are limited to one task.
Approach:	They propose to evaluate Chinese language models' biases from multiple perspectives using a multi-task Chinese Bias Evaluation Benchmark.
Outcome:	The proposed model covers 12, 82 subcategories and 5 evaluation tasks covering a wide range of categories and content diversity.

Agent-Pro: Learning to Evolve via Policy-Level Reflection and Optimization (2024.acl-long)

Copied to clipboard

Wenqi Zhang, Ke Tang, Hai Wu, Mengna Wang, Yongliang Shen, Guiyang Hou, Zeqi Tan, Peng Li, Yueting Zhuang, Weiming Lu

RAG-Instruct: Boosting LLMs with Diverse Retrieval-Augmented Instructions (2025.emnlp-main)

Challenge:	Large Language Models (LLMs) are designed as specific task solvers with sophisticated prompt engineering, but are inherently incapacitating to address complex dynamic scenarios.
Approach:	They propose an LLM-based agent with policy-level reflection and optimization that can learn from interactive experiences and progressively elevate its behavioral policy.
Outcome:	The proposed agent outperforms vanilla LLM and specialized models in blackjack and Texas hold’em.

Copied to clipboard

Wanlong Liu, Junying Chen, Ke Ji, Li Zhou, Wenyu Chen, Benyou Wang

PROSE: A Pronoun Omission Solution for Chinese-English Spoken Language Translation (2023.emnlp-main)

Challenge:	Retrieval-Augmented Generation (RAG) has emerged as a key paradigm for enhancing large language models by incorporating external knowledge.
Approach:	They propose a method for synthesizing diverse and high-quality RAG instruction data based on any source corpus.
Outcome:	The proposed method outperforms existing methods in multiple tasks and achieves strong zero-shot performance.

Copied to clipboard

Ke Wang, Xiutian Zhao, Yanghui Li, Wei Peng

Machine Translation

Challenge:	Pro-drop (‘pronoun-dropping’) language requires NMT systems to recover omitted pronouns, but this task lacks sufficient datasets for benchmarking .
Approach:	They propose a benchmarking method that leverages the semantic embedding of dropped pronouns to augment training pairs to alleviate the negative impact introduced by pro-drop .
Outcome:	The proposed method outperforms existing methods regarding omitted pronoun retrieval and overall translation quality on four Chinese-English translation corpora.

Probability-Consistent Preference Optimization for Enhanced LLM Reasoning (2025.findings-acl)

Copied to clipboard

Yunqiao Yang, Houxing Ren, Zimu Lu, Ke Wang, Weikang Shi, Aojun Zhou, Junting Pan, Mingjie Zhan, Hongsheng Li

InferAligner: Inference-Time Alignment for Harmlessness through Cross-Model Guidance (2024.emnlp-main)

Challenge:	Recent advances in preference optimization have demonstrated significant potential for improving mathematical reasoning capabilities in large language models.
Approach:	They propose a framework that establishes two quantitative metrics for preference selection: surface-level answer correctness and intrinsic token-level probability consistency.
Outcome:	The proposed framework outperforms existing outcome-only criterion approaches across a diverse range of LLMs and benchmarks.

Copied to clipboard

Pengyu Wang, Dong Zhang, Linyang Li, Chenkun Tan, Xinghao Wang, Mozhi Zhang, Ke Ren, Botian Jiang, Xipeng Qiu

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

Challenge:	Existing methods for enhancing harmlessness and helpfulness of large language models (LLMs) involve complex and resource-intensive training processes.
Approach:	They propose a method that decouples harmlessness from helpfulness during inference phase.
Outcome:	The proposed method significantly reduces the attack success rate (ASR) of harmful instructions and jailbreak instructions while maintaining almost unchanged performance in downstream tasks.

Copied to clipboard

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

HybridKV: Hybrid KV Cache Compression for Efficient Multimodal Large Language Model Inference (2026.acl-long)

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

Bowen Zeng, Feiyang Ren, Jun Zhang, Xiaoling Gu, Ke Chen, Lidan Shou, Huan Li

Training Language Model to Critique for Better Refinement (2025.findings-acl)

Challenge:	Multimodal Large Language Models (MLLMs) are hindered by the rapid growth of key–value (KV) caches.
Approach:	They propose a hybrid KV cache compression framework that reduces KV memory by 7.9 and speeds up decoding by 1.52.
Outcome:	Experiments on 11 multimodal benchmarks show that HYBRIDKV cuts KV cache memory by 7.9 and speeds up decoding by 1.52.

Copied to clipboard

Tianshu Yu, Chao Xiang, Mingchuan Yang, Pei Ke, Bosi Wen, Cunxiang Wang, Jiale Cheng, Li Zhang, Xinyu Mu, Chuxiong Sun, Minlie Huang

Speculating LLMs’ Chinese Training Data Pollution from Their Tokens (2025.emnlp-main)

Challenge:	Large language models (LLMs) have remarkable evaluation and critique capabilities, providing insightful feedback and identifying flaws in various tasks.
Approach:	They propose a framework to train critic models using refinement signals to generate feedback loops where critiques guide the model in refining its responses.
Outcome:	The proposed framework outperforms traditional methods and open-source models in terms of critique quality and refinement outcomes.

Copied to clipboard

Qingjie Zhang, Di Wang, Haoting Qian, Liu Yan, Tianwei Zhang, Ke Xu, Qi Li, Minlie Huang, Hewu Li, Han Qiu

InstOptima: Evolutionary Multi-objective Instruction Optimization via Large Language Model-based Instruction Operators (2023.findings-emnlp)

Challenge:	Experiments on GPT and other 23 LLMs indicate that tokens widely exist while GPT’s vocabulary behaves the worst: more than 23% long Chinese tokens (i.e., a token with more than two Chinese characters) are either porn or online gambling.
Approach:	They propose to locate Polluted Chinese (PoC) tokens in LLMs and build a PoC token detector to label them in vocabularies by considering each token’s semantics and related contents from the search engines.
Outcome:	The proposed method predicts that the ratio of “*” related webpages in GPT-4o's training data is around 0.5%.

Copied to clipboard

PaTaRM: Bridging Pairwise and Pointwise Signals via Preference-Aware Task-Adaptive Reward Modeling (2026.acl-long)

Challenge:	Existing studies focus on automating instruction generation but do not consider other objectives that impact instruction quality.
Approach:	They propose an approach that treats instruction generation as an evolutionary multi-objective optimization problem.
Outcome:	The proposed approach improves fine-tuning performance and the generation of high-quality instructions.

Copied to clipboard

Ai Jian, Jingqing Ruan, Xing Ma, Dailin Li, Weipeng Zhang, Ke Zeng, Xunliang Cai

Challenge:	Existing reward models lack generative and reasoning capabilities, resulting in poor performance.
Approach:	They propose a reward-aware task-adaptive reward model that enables pointwise training using readily available pairwise data via a novel Preference-Aware Reward mechanism.
Outcome:	The proposed reward model achieves an average relative improvement of 8.7% over the base models on RewardBench and RMBench.

MathCanvas: Intrinsic Visual Chain-of-Thought for Multimodal Mathematical Reasoning (2026.acl-long)

Copied to clipboard

Weikang Shi, Aldrich Yu, Rongyao Fang, Houxing Ren, Ke Wang, Aojun Zhou, Changyao Tian, Xinyu Fu, Yuxuan Hu, Zimu Lu, Linjiang Huang, Si Liu, Rui Liu, Hongsheng Li

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

Challenge:	Existing approaches to visual chain-of-thought are limited by external tools or fail to generate high-fidelity diagrams.
Approach:	They propose a framework to enable large multimodal models with VCoT capabilities . they pre-train a model on a 15.2M-pair corpus and teach it how to leverage visual aids .
Outcome:	The proposed framework unlocks complex, human-like visual reasoning in large language models . it pre-trains the model on a 15.2M-pair corpus and fine-tunes it on MathCanvas-Instruct .

Copied to clipboard

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

Comprehensive Benchmarking of Long-Form Speech Generation in Diverse Scenarios (2026.findings-acl)

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

Changhao Pan, Rui Yang, Han Wang, Zhuan Zhou, Xuming He, Wenxiang Guo, Ziyue Jiang, Ruiqi Li, Yu Zhang, Chenyuhao Wen, Ke Lei, Xiang Yin, Jingyu Lu, Zhiyuan Zhu, Zhou Zhao

SpeechIQ: Speech-Agentic Intelligence Quotient Across Cognitive Levels in Voice Understanding by Large Language Models (2025.acl-long)

Challenge:	Existing evaluation benchmarks for long-form speech are limited to limited domains, creating a significant gap with the diverse downstream applications.
Approach:	They propose a benchmark that decomposes "long-form speech quality" into specific, disentangled dimensions.
Outcome:	The proposed benchmark decomposes “long-form speech quality” into specific, disentangled dimensions.

Copied to clipboard

Zhen Wan, Chao-Han Huck Yang, Yahan Yu, Jinchuan Tian, Sheng Li, Ke Hu, Zhehuai Chen, Shinji Watanabe, Fei Cheng, Chenhui Chu, Sadao Kurohashi

Beyond Templates: Dynamic Adaptation of Reasoning Demonstrations via Feasibility-Aware Exploration (2026.findings-acl)

Challenge:	SIQ quantifies voice understanding abilities and provides unified comparisons between cascaded methods and end-to-end models.
Approach:	They propose a human cognition-inspired evaluation pipeline for voice understanding large language models (LLM_Voice) that quantifies voice understanding abilities and provides unified comparisons between cascaded methods and end-to-end models.
Outcome:	The proposed framework quantifies voice understanding abilities and provides unified comparisons between cascaded methods and end-to-end models, identifies annotation errors in existing benchmarks, and detects hallucinations in LLM_Voice.

Copied to clipboard

Yong Wu, Weihang Pan, Ke Li, Chen Binhui, Ping Li, Binbin Lin

C3KG: A Chinese Commonsense Conversation Knowledge Graph (2022.findings-acl)

Challenge:	Existing reasoning datasets that are designed for powerful LLMs often lead to degraded performance when directly applied to weaker models.
Approach:	They propose a data adaptation framework that bridges the capability gap between expert reasoning trajectories and diverse SLMs by employing a selective imitation strategy guided by step-wise adaptability estimation via solution simulation.
Outcome:	The proposed framework improves generalization and data efficiency over static fine-tuning and can be applied to large models with limited model capacity.

Copied to clipboard

Dawei Li, Yanran Li, Jiayi Zhang, Ke Li, Chen Wei, Jianwei Cui, Bin Wang

TopWORDS-Poetry: Simultaneous Text Segmentation and Word Discovery for Classical Chinese Poetry via Bayesian Inference (2023.emnlp-main)

Challenge:	Existing commonsense knowledge bases organize tuples in an isolated manner, causing problems for chatbots .
Approach:	They create a Chinese commonsense conversation knowledge graph which integrates social commonsensm and dialog flow information.
Outcome:	The proposed graph incorporates social commonsense knowledge and dialog flow information.

Copied to clipboard

Changzai Pan, Feiyue Li, Ke Deng

Syntactic Text Processing

Challenge:	Experimental studies confirm that TopWORDS-Poetry can successfully segment poetry words without pre-given vocabulary or training corpus.
Approach:	They propose an unsupervised method that can achieve reliable text segmentation and word discovery for classical Chinese poetry simultaneously without pre-given vocabulary or training corpus.
Outcome:	Experimental results show that TopWORDS-Poetry can segment poetry lines into meaningful words with high quality without pre-given vocabulary or training corpus.

MdEval: Massively Multilingual Code Debugging (2026.findings-acl)

Copied to clipboard

Shukai Liu, Linzheng Chai, Jian Yang, Jiajun Shi, He Zhu, Liran Wang, Jin Ke, Wei Zhang, Hualei Zhu, Shuyue Guo, Tao Sun, Jiaheng Liu, Yunlong Duan, Yu Hao, Liqun Yang, Guanglin Niu, Ge Zhang, Zhoujun Li

GigaSpeech 2: An Evolving, Large-Scale and Multi-domain ASR Corpus for Low-Resource Languages with Automated Crawling, Transcription and Refinement (2025.acl-long)

Challenge:	Existing benchmarks primarily focus on Python and are limited in terms of language diversity.
Approach:	They propose a multilingual debugging benchmark that includes 3.9K test samples of 20 programming languages and introduces the debug instruction corpora MdEval-Instruct by injecting bugs into the correct multilingual queries and solutions.
Outcome:	The proposed benchmark includes 3.9K test samples of 20 programming languages and covers the automated program repair task, bug localization task, and bug identification task.

Copied to clipboard

Yifan Yang, Zheshu Song, Jianheng Zhuo, Mingyu Cui, Jinpeng Li, Bo Yang, Yexing Du, Ziyang Ma, Xunying Liu, Ziyuan Wang, Ke Li, Shuai Fan, Kai Yu, Wei-Qiang Zhang, Guoguo Chen, Xie Chen

Sinkhorn Distance Minimization for Knowledge Distillation (2024.lrec-main)

Challenge:	GigaSpeech 2 is a large-scale, multi-domain, multilingual speech recognition corpus for low-resource languages.
Approach:	They propose a large-scale, multi-domain, multilingual speech recognition corpus for low-resource languages and an automated pipeline for data crawling, transcription, and label refinement.
Outcome:	The proposed corpus reduces the word error rate for Thai, Indonesian, and Vietnamese on a realistic YouTube test set by 25% to 40% compared to Whisper large-v3.

Copied to clipboard

Xiao Cui, Yulei Qin, Yuting Gao, Enwei Zhang, Zihan Xu, Tong Wu, Ke Li, Xing Sun, Wengang Zhou, Houqiang Li

EPO: Explicit Policy Optimization for Strategic Reasoning in LLMs via Reinforcement Learning (2025.acl-long)

Challenge:	Existing knowledge distillation methods investigate divergence measures but fail to deliver effective supervision when few distribution overlap exists between teacher and student.
Approach:	They propose a knowledge distillation method that exploits the Sinkhorn distance to ensure a nuanced assessment of the disparity between teacher and student distributions.
Outcome:	The proposed method outperforms state-of-the-art methods on all kinds of LLMs with encoder-only, encoder decoder, and decoded architectures.

Copied to clipboard

Xiaoqian Liu, Ke Wang, Yongbin Li, Yuchuan Wu, Wentao Ma, Aobo Kong, Fei Huang, Jianbin Jiao, Junge Zhang

LM-Searcher: Cross-domain Neural Architecture Search with LLMs via Unified Numerical Encoding (2025.emnlp-main)

Challenge:	Existing methods for strategic reasoning face challenges in adaptability, scalability, and transferring strategies to new contexts.
Approach:	They propose an explicit policy optimization model that provides strategies in open-ended action space and can be plugged into arbitrary LLM agents to motivate goal-directed behavior.
Outcome:	The proposed model provides strategies in open-ended action space and can be plugged into arbitrary LLM agents to motivate goal-directed behavior.

Copied to clipboard

Yuxuan Hu, Jihao Liu, Ke Wang, Jinliang Zheng, Weikang Shi, Manyuan Zhang, Qi Dou, Rui Liu, Aojun Zhou, Hongsheng Li

Understanding the Dark Side of LLMs’ Intrinsic Self-Correction (2025.acl-long)

Challenge:	Recent advances in Large Language Models have opened new avenues for solving complex optimization problems, including Neural Architecture Search (NAS).
Approach:	They propose a framework that leverages LLMs for cross-domain neural architecture optimization without extensive domain-specific tuning.
Outcome:	The proposed framework achieves competitive performance in both in-domain and out-of-domain tasks.

Copied to clipboard

Qingjie Zhang, Di Wang, Haoting Qian, Yiming Li, Tianwei Zhang, Minlie Huang, Ke Xu, Hewu Li, Liu Yan, Han Qiu

Challenge:	Recent studies show that LLMs’ intrinsic self-correction fails without oracle labels as feedback.
Approach:	They propose to use one simple task and three complex tasks with state-of-the-art LLMs like ChatGPT, Llama, and DeepSeek to interpret LLM's intrinsic self-correction.
Outcome:	The proposed methods reveal the dark side of LLMs’ intrinsic self-correction for different tasks, especially for those failure cases.

The Best Defense is Attack: Repairing Semantics in Textual Adversarial Examples (2024.emnlp-main)

Copied to clipboard

SDPO: Segment-Level Direct Preference Optimization for Social Agents (2025.acl-long)

Robustness in NLP

Challenge:	Recent studies have revealed the vulnerability of pre-trained language models to adversarial attacks.
Approach:	They propose a novel approach to repair adversarial examples using an adversarial detector.
Outcome:	The proposed approach is effective in various adversarial attack scenarios.

Copied to clipboard

Aobo Kong, Wentao Ma, Shiwan Zhao, Yongbin Li, Yuchuan Wu, Ke Wang, Xiaoqian Liu, Qicheng Li, Yong Qin, Fei Huang

Challenge:	Direct Preference Optimization (DPO) has proven effective in aligning LLM behavior with human preferences across various tasks, but is limited in multi-turn social interactions.
Approach:	They propose a method which dynamically selects key segments within interactions to optimize multi-turn agent behavior.
Outcome:	The proposed methods outperform existing methods and proprietary LLMs on the SOTOPIA benchmark and show that they can improve social intelligence.

PAMN: Multi-phase Correlation Modeling for Contrast-Enhanced 3D Medical Image Retrieval (2025.findings-emnlp)

Copied to clipboard

Haonan Tong, Ke Liu, Chuang Zhang, Xinglin Zhang, Tao Chen, Jenq-Neng Hwang, Lei Li

SEC-FinTables: Evaluating Large Language Models for Detecting Logical Inconsistencies on Tabular Data (2026.findings-acl)

Challenge:	Current 3D medical imaging models focus on spatial features, neglecting phase-specific progression detailed in clinical reports.
Approach:	They propose a framework that fuses imaging phases with clinical text to enhance 3D medical image retrieval.
Outcome:	The proposed framework outperforms state-of-the-art models on a phase-series dataset of 12,230 hospital CT scans.

Copied to clipboard

Shuyan Ke, Qiong Wu, Hui Li, Liujuan Cao

QiMeng-Attention: SOTA Attention Operator is generated by SOTA Attention Algorithm (2025.findings-acl)

Challenge:	Large language models are increasingly deployed in high-stakes domains where logical inconsistencies are unrecognized.
Approach:	They propose a benchmarking system that decomposes inconsistency detection into granular subtasks and a protocol that decompiles it into subtask.
Outcome:	The proposed model decomposes inconsistencies into subtasks and identifies them in 103,395 real-world and error-injected table instances.

Copied to clipboard

Qirui Zhou, Shaohui Peng, Weiqiang Xiong, Haixin Chen, Yuanbo Wen, Haochen Li, Ling Li, Qi Guo, Yongwei Zhao, Ke Gao, Ruizhi Chen, Yanjun Wu, Zhao Chen, Yunji Chen

Programming Languages in NLP

Challenge:	Existing LLMs cannot comprehend the complex data flow and computation process of the attention operator and utilize low-level primitive to exploit GPU performance.
Approach:	They propose an LLM-friendly Thinking Language (LLM-TL) that can decouple the generation of high-level optimization logic and low-level implementation on GPU and enhance LLMs’ understanding of attention operator.
Outcome:	The proposed method outshines existing LLMs on A100, RTX8000, and T4 GPUs, achieving a speed-up of up to 35.16.

MIO: A Foundation Model on Multimodal Tokens (2025.emnlp-main)

Copied to clipboard

Zekun Moore Wang, King Zhu, Chunpu Xu, Wangchunshu Zhou, Jiaheng Liu, Yibo Zhang, Jessie Wang, Ning Shi, Siyu Li, Yizhi Li, Haoran Que, Zhaoxiang Zhang, Yuanxing Zhang, Ge Zhang, Ke Xu, Jie Fu, Wenhao Huang

Watch Every Step! LLM Agent Learning via Iterative Step-level Process Refinement (2024.emnlp-main)

Challenge:	Existing models lack multimodal understanding capabilities, resulting in closed-source model that does not support multimodal interleaved sequences.
Approach:	They propose a foundation model built on multimodal tokens capable of understanding and generating speech, text, images, and videos in an end-to-end, autoregressive manner.
Outcome:	The proposed model is able to understand speech, text, images, and videos in an end-to-end, autoregressive manner.

Copied to clipboard

Weimin Xiong, Yifan Song, Xiutian Zhao, Wenhao Wu, Xun Wang, Ke Wang, Cheng Li, Wei Peng, Sujian Li

Investigating Learning Dynamics of BERT Fine-Tuning (2020.aacl-main)

Challenge:	Recent approaches to enhance agent performance focus on outcome rewards, which may lead to errors or suboptimal actions due to the absence of process supervision signals.
Approach:	They propose a step-level framework that provides detailed step-by-step guidance to enhance agent training by using Monte Carlo methods.
Outcome:	The proposed framework outperforms strong baselines on three tasks and shows that it is effective in augmenting efficiency and its applicability to diverse models.

Copied to clipboard

Yaru Hao, Li Dong, Furu Wei, Ke Xu

Alignment with Fill-In-the-Middle for Enhancing Code Generation (2025.emnlp-main)

Challenge:	Recent studies have shown that the fine-tuning process improves performance on downstream tasks.
Approach:	They propose two new pre-training tasks to improve the model performance on downstream tasks.
Outcome:	The proposed model achieves state-of-the-art on a wide array of NLP tasks.

Copied to clipboard

Houxing Ren, Zimu Lu, Weikang Shi, Haotian Hou, Yunqiao Yang, Ke Wang, Aojun Zhou, Junting Pan, Mingjie Zhan, Hongsheng Li

Programming Languages in NLP

Challenge:	Existing methods for generating test cases with limited training data are not reliable and may be counterproductive.
Approach:	They propose a method that splits code snippets into smaller, granular blocks, creating more diverse DPO pairs from the same test cases.
Outcome:	The proposed approach shows significant improvements in code generation tasks on benchmark datasets such as HumanEval (+), MBPP (+), and APPS.

KG-ViP: Bridging Knowledge Grounding and Visual Perception in Multi-modal LLMs for Visual Question Answering (2026.acl-long)

Copied to clipboard

Zhiyang Li, Ao Ke, Yukun Cao, Xike Xie

Enhancing the General Agent Capabilities of Low-Paramter LLMs through Tuning and Multi-Branch Reasoning (2024.findings-naacl)

Challenge:	Existing approaches to Visual Question Answering lack synergistic potential of scene graphs and scene graph.
Approach:	They propose a retrieval-and-fusion pipeline that fuses scene graphs and commonsense graphs to enable multi-modal reasoning.
Outcome:	Experiments on FVQA 2.0+ and MVQA benchmarks show that KG-ViP outperforms existing methods.

Copied to clipboard

Qinhao Zhou, Zihan Zhang, Xiang Xiang, Ke Wang, Yuchuan Wu, Yongbin Li

Towards Robust Real-World Spreadsheet Understanding with Multi-Agent Multi-Format Reasoning (2026.acl-long)

Challenge:	Open-source pre-trained Large Language Models exhibit strong language understanding and generation capabilities, making them highly successful in a variety of tasks.
Approach:	They propose a method to construct agent-specific data using GPT-4 and supervised fine-tuning . they find that supervised tunning can significantly reduce hallucination outputs and formatting errors in agent tasks .
Outcome:	The proposed method improves on five agent tasks of AgentBench.

Copied to clipboard

Houxing Ren, Mingjie Zhan, Zimu Lu, Ke Wang, Yunqiao Yang, Haotian Hou, Hongsheng Li

Robustness in NLP

Challenge:	Spreadsheets are among the most widely used data formats in real-world applications . existing large language models treat tables as plain text, overlooking layout cues and visual semantics.
Approach:	They propose a two-stage multi-agent framework for spreadsheet understanding that adopts a step-by-step reading and reasoning paradigm.
Outcome:	Extensive experiments on two spreadsheet datasets show the proposed framework outperforms existing methods on Spreadsheet Bench.

When to Continue Thinking: Adaptive Thinking Mode Switching for Efficient Reasoning (2025.findings-emnlp)

Copied to clipboard

Xiaoyun Zhang, Jingqing Ruan, Xing Ma, Yawen Zhu, Haodong Zhao, Hao Li, Jiansong Chen, Ke Zeng, Xunliang Cai

Responsible & Trustworthy NLP

Challenge:	Large reasoning models (LRMs) incur excessive computational overhead due to redundant reasoning, especially on simple tasks.
Approach:	They propose an Adaptive Self-Recovery Reasoning framework that suppresses unnecessary reasoning and enables implicit recovery.
Outcome:	The proposed framework suppresses unnecessary reasoning and enables implicit recovery.

LLM-Guided Semantic-Aware Clustering for Topic Modeling (2025.acl-long)

Copied to clipboard

Jianghan Liu, Ziyu Shang, Wenjun Ke, Peng Wang, Zhizhao Luo, Jiajun Liu, Guozheng Li, Yining Li

Information Extraction & Text Mining

Challenge:	Experimental results show that topic modeling is competitive compared to closed-source methods.
Approach:	They propose a semi-supervised topic modeling method that combines LLMs with clustering to improve topic generation and distribution.
Outcome:	The proposed method outperforms state-of-the-art methods that utilize GPT-4 on topic alignment and exhibits competitive performance compared to Neural Topic Models on topic quality.

PersonaLM: Language Model Personalization via Domain-distributed Span Aggregated K-Nearest N-gram Retrieval Augmentation (2023.findings-emnlp)

Copied to clipboard

Puneet Mathur, Zhe Liu, Ke Li, Yingyi Ma, Gil Keren, Zeeshan Ahmed, Dinesh Manocha, Xuedong Zhang

Boosting Textural NER with Synthetic Image and Instructive Alignment (2024.findings-acl)

Challenge:	Existing language modeling tools for automatic speech recognition (ASR) are difficult to personalize.
Approach:	They propose a domain-distributed Span-Aggregated K-nearest N-gram retrieval augmentation to improve language modeling for automatic speech recognition (ASR) personalization.
Outcome:	The proposed model outperforms baselines on Wikitext-103, UserLibri, and ASAP datasets with a 10-16% improvement in perplexity and a 5-8% reduction in word error rates.

Copied to clipboard

Jiahao Wang, Wenjun Ke, Peng Wang, Hang Zhang, Dong Nie, Jiajun Liu, Guozheng Li, Ziyu Shang

Boosting Text Augmentation via Hybrid Instance Filtering Framework (2023.findings-acl)

Challenge:	Named entity recognition (NER) is a key task reliant on textual data.
Approach:	They propose a method to transform NER into a multimodal task by using images from the internet as auxiliaries.
Outcome:	The proposed method surpasses all text-only baselines and improves F1 score by 1.4% to 2.3% on prominent MNER datasets.

Copied to clipboard

Adaption-of-Thought: Learning Question Difficulty Improves Large Language Models for Reasoning (2024.emnlp-main)

Challenge:	Existing text augmentation methods generate instances with shifted feature spaces, which leads to a drop in performance on large datasets.
Approach:	They propose a hybrid instance-filtering framework that generates instances with shifted feature spaces, which leads to a drop in performance on augmented data.
Outcome:	The proposed framework outperforms state-of-the-art methods on three classification tasks and nine public datasets.

Copied to clipboard

Mayi Xu, Yongqi Li, Ke Sun, Tieyun Qian

Fin-STAR: Structure-as-Semantics to Resolve Implicitness in Financial Retrieval (2026.findings-acl)

Challenge:	Existing methods do not differentiate question difficulty when designing prompting methods for them.
Approach:	They propose an adaptive method to improve large language models for reasoning problems by measuring question difficulty and tailoring demonstration set construction and difficulty-adapted retrieval strategies.
Outcome:	The proposed method shows an absolute improvement of up to 5.5% on arithmetic reasoning, 7.4% on symbolic reasoning, and 2.3% on commonsense reasoning.

Copied to clipboard

Yu Zou, Yan Chen, Lida He, Qi Zhou, Xiaorui Zhou, Aixi Zhong, Yi Wang, Wei Li, Qingyu Wang, Jiatao Li, Wei Gong, Jialei Zeng, Jingmei Zhao, Ke Jiang, Qing Li

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

Challenge:	Existing Retrieval-Augmented Generation systems treat structure as a physical navigational skeleton rather than intrinsic semantic knowledge.
Approach:	They propose a framework that redefining hierarchy as intrinsic semantics and uses snippets to enrich hierarchical lineage.
Outcome:	The proposed framework outperforms state-of-the-art hierarchical and graph-based benchmarks on FinTierQA Gold.

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

Green & Sustainable NLP

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.

PerturbScore: Connecting Discrete and Continuous Perturbations in NLP (2023.findings-emnlp)

Copied to clipboard

Linyang Li, Ke Ren, Yunfan Shao, Pengyu Wang, Xipeng Qiu

Robustness in NLP

Challenge:	Natural language processing (NLP) applications are growing rapidly due to discrete nature of texts.
Approach:	They propose to connect discrete perturbations with continuous perturbations to help understand discrete ones in NLP models.
Outcome:	The proposed method surpasses methods used in discrete perturbation measuring and can be generalized to different datasets, perturbation methods.

Modeling Aspect Sentiment Coherency via Local Sentiment Aggregation (2024.findings-eacl)

Copied to clipboard

Information Extraction & Text Mining

Challenge:	Existing studies have not explored aspect sentiment coherency, including its implications in adversarial defense.
Approach:	They propose a local sentiment aggregation paradigm that models aspect sentiment coherency . they demonstrate the capability of LSA in adversarial defense .
Outcome:	The proposed model outperforms existing models and achieves state-of-the-art sentiment classification performance.

MOA: Multi-Objective Alignment for Role-Playing Agents (2026.acl-long)

Copied to clipboard

Chonghua Liao, Ke Wang, Yuchuan Wu, Ruoran Li, Fei Huang, Yongbin Li

Challenge:	Prior work on role-playing agents relies on supervised fine-tuning or reinforcement learning with scalarized rewards, but these approaches do not address the coordination of multiple reward dimensions during optimization.
Approach:	They propose a reinforcement-learning framework that enables multi-dimensional, fine-grained rubric optimization for general RPAs.
Outcome:	Experiments on PersonaGym and RoleMRC show that MOA improves multi-dimensional role-playing performance over supervised and standard RL baselines.

DL-QAT: Weight-Decomposed Low-Rank Quantization-Aware Training for Large Language Models (2024.emnlp-industry)

Copied to clipboard

Wenjing Ke, Zhe Li, Dong Li, Lu Tian, Emad Barsoum

Flexibly Utilize Memory for Long-Term Conversation via a Fragment-then-Compose Framework (2025.emnlp-main)

Challenge:	Quantization-aware Training (QAT) is a popular technique for reducing memory usage and improving computational efficiency in large language models.
Approach:	They propose a weight-decomposed low-rank quantization-aware training approach that integrates QAT with a group-specific quantization magnitude adjustment.
Outcome:	The proposed method outperforms the state-of-the-art method on LLaMA and LLama2 models.

Copied to clipboard

Cai Ke, Yiming Du, Bin Liang, Yifan Xiang, Lin Gui, Zhongyang Li, Baojun Wang, Yue Yu, Hui Wang, Kam-Fai Wong, Ruifeng Xu

Dialogue Systems & Conversational Agents

Challenge:	Large language models extract useful information from conversation history to enhance the response in long-term conversations.
Approach:	They propose a Fragment-then-Compose framework to optimize memory utilization for long-term open-domain conversation.
Outcome:	The proposed framework can be used to extract useful information from conversation history . it can be adapted to different situations and improve response generation .

Amadeus: Autoregressive Model with Bidirectional Attribute Modelling for Symbolic Music (2026.acl-long)

Copied to clipboard

Hongju Su, Ke Li, Lan Yang, Honggang Zhang, Yi-Zhe Song

DPEPO: Diverse Parallel Exploration Policy Optimization for LLM-based Agents (2026.acl-long)

Challenge:	Existing symbolic music generation models represent musical notes as a sequence of attribute tokens with fixed unidirectional dependencies.
Approach:	They propose a symbolic music generation framework that adopts a autoregressive and a discrete diffusion architectures for note attributes.
Outcome:	The proposed framework improves state-of-the-art models across objective and subjective metrics.

Copied to clipboard

JunShuo Zhang, Chengrui Huang, Feng Guo, Zihan Li, Ke Shi, Menghua Jiang, Jiguo Yu, Shuo Shang, Shen Gao

Unveiling the Implicit Toxicity in Large Language Models (2023.emnlp-main)

Challenge:	Existing approaches to large language model (LLM) agents that follow the sequential "reason-then-act" paradigm suffer from limited exploration and incomplete environmental understanding as they interact with only a single environment per step.
Approach:	They propose a paradigm that enables an agent to interact with multiple environments simultaneously and share cross-trajectory experiences.
Outcome:	The proposed paradigm achieves state-of-the-art (SOTA) success rates while maintaining comparable efficiency to strong sequential baselines.

Copied to clipboard

Jiaxin Wen, Pei Ke, Hao Sun, Zhexin Zhang, Chengfei Li, Jinfeng Bai, Minlie Huang

Improving BERT with Syntax-aware Local Attention (2021.findings-acl)

Challenge:	Recent studies focus on probing toxic outputs that can be easily detected with existing toxicity classifiers, but LLMs can generate diverse implicit toxic output that are difficult to detect via simply zero-shot prompting.
Approach:	They propose a reinforcement learning based attacking method to induce the implicit toxic outputs in large language models by fine-tuning toxicity classifiers.
Outcome:	The proposed method generates implicit toxic outputs that are difficult to detect via zero-shot prompting on five widely-adopted toxicity classifiers.

Copied to clipboard

Zhongli Li, Qingyu Zhou, Chao Li, Ke Xu, Yunbo Cao

DuSQL: A Large-Scale and Pragmatic Chinese Text-to-SQL Dataset (2020.emnlp-main)

Challenge:	Recent studies show that attention-based models benefit from more focused attention over local regions.
Approach:	They propose a syntax-aware local attention which restrains attention over syntactically relevant words.
Outcome:	The proposed model performs better on all benchmark datasets, including sentence classification and sequence labeling tasks.

Copied to clipboard

Lijie Wang, Ao Zhang, Kun Wu, Ke Sun, Zhenghua Li, Hua Wu, Min Zhang, Haifeng Wang

Programming Languages in NLP

Challenge:	Existing text-to-SQL parsing methods mainly focus on English, but there is no labeled data available for the language . a larges-scale and pragmatic Chinese dataset is used for cross-domain text- to-Sql task .
Approach:	They propose a larges-scale Chinese dataset for a cross-domain text-to-SQL task . they analyze questions from several representative applications and use an effective data construction framework .
Outcome:	The proposed dataset contains 200 databases, 813 tables, and 23,797 question/SQL pairs.

AudioChatLlama: Towards General-Purpose Speech Abilities for LLMs (2024.naacl-long)

Copied to clipboard

Yassir Fathullah, Chunyang Wu, Egor Lakomkin, Ke Li, Junteng Jia, Yuan Shangguan, Jay Mahadeokar, Ozlem Kalinli, Christian Fuegen, Mike Seltzer

Learning to Sample Replacements for ELECTRA Pre-Training (2021.findings-acl)

Challenge:	a new model for speech processing and reasoning uses curated data instead of text.
Approach:	They extend the instruction-tuned Llama-2 model with end-to-end speech processing and reasoning abilities without using any carefully curated paired data.
Outcome:	The proposed model outperforms or outperfects existing models on synthesized and recorded speech QA tests.

Copied to clipboard

Yaru Hao, Li Dong, Hangbo Bao, Ke Xu, Furu Wei

Entriever: Energy-based Retriever for Knowledge-Grounded Dialog Systems (2025.findings-acl)

Challenge:	Experimental results show that ELECTRA pretrains a discriminator to detect replaced tokens . despite compelling performance, there is no direct feedback loop from discriminator and generator to generator, making replacements biased to correct tokens.
Approach:	They propose to augment sampling with a hardness prediction mechanism to encourage the discriminator to learn what it has not acquired.
Outcome:	The proposed method improves ELECTRA pre-training on various downstream tasks.

Copied to clipboard

Yucheng Cai, Ke Li, Yi Huang, Junlan Feng, Zhijian Ou

Unlocking Instructive In-Context Learning with Tabular Prompting for Relational Triple Extraction (2024.lrec-main)

Challenge:	Existing knowledge retrieval methods fail to account for interrelationship between knowledge pieces . however, current methods fail in a situation where multiple knowledge pieces are relevant .
Approach:	They propose an energy-based retriever that directly models the candidate retrieval results as a whole instead of modeling the knowledge pieces separately.
Outcome:	The proposed retriever outperforms the baseline energy-based retriever in knowledge retrieval tasks.

Copied to clipboard

Guozheng Li, Wenjun Ke, Peng Wang, Zijie Xu, Ke Ji, Jiajun Liu, Ziyu Shang, Qiqing Luo

Visualizing and Understanding the Effectiveness of BERT (D19-1)

Challenge:	Existing methods for relational triple extraction (RTE) are unnatural and recast RTE tasks to text-to-text prompting formats.
Approach:	They propose a tabular prompting for RTE which frames RTE task into a table generation task and propose an instructive in-context learning which only selects and annotates samples considering triple semantics in massive unlabeled samples.
Outcome:	The proposed prompting for RTE with TableIE achieves state-of-the-art performance compared to other methods . the proposed prompts are based on off-the shelf LLMs and are scalable to multiple scenarios .

Copied to clipboard

Yaru Hao, Li Dong, Furu Wei, Ke Xu

FlowBench: Revisiting and Benchmarking Workflow-Guided Planning for LLM-based Agents (2024.findings-emnlp)

Challenge:	Language model pre-training, such as BERT, has achieved strong performance in many NLP tasks.
Approach:	They propose to visualize loss landscapes and optimization trajectories of fine-tuning BERT on specific datasets.
Outcome:	The proposed model improves performance and generalization capability across tasks.

Copied to clipboard

Ruixuan Xiao, Wentao Ma, Ke Wang, Yuchuan Wu, Junbo Zhao, Haobo Wang, Fei Huang, Yongbin Li

Challenge:	LLM-based agents are susceptible to undesired planning hallucinations when lacking specific knowledge for expertise-intensive tasks.
Approach:	They propose a benchmark to evaluate the efficacy of workflow-guided agent planning by formalizing different formats of workflow knowledge.
Outcome:	The proposed benchmark aims to improve the planning reliability of LLM-based agents by incorporating external workflow-related knowledge.

Relaxing the Constraints: A Dual-Importance Projection Mechanism for Lifelong Model Editing (2026.findings-acl)

Copied to clipboard

Zhenghai Chen, Senbin Xu, Jiaxi Tan, Xinhua Wu, Yan Zhang, Xiawu Zheng, Shengchuan Zhang, Ke Li, Sicheng Zhao, Liujuan Cao, Rongrong Ji

EviReport: From Reasoned Outlines to Evidence Tracked Long-Form Reports (2026.findings-acl)

Challenge:	Existing knowledge editing methods rely on strict orthogonal projection to preserve previously edited knowledge, but this constraint limits gradient expressiveness, resulting in degradation of model generalization and overall performance as the number of edits increases.
Approach:	They propose a method that leverages Singular Value Decomposition to identify critical gradient subspaces and introduces a dual mechanism comprising "accumulated importance" and "projection importance"
Outcome:	Extensive experiments on five mainstream LLMs show that the proposed method achieves an average comprehensive performance improvement of 10.36% and effectively maintains the model’s general capabilities on downstream tasks.

Copied to clipboard

Zihan Liu, Jianhui li, Zexin Wang, Fei Sun, Jingjing LI, Zheyuan Li, Ke Xiang, Hang Cui, Houhua Gong, Changhua Pei, Gaogang Xie

ProtoCycle: Reflective Tool-Augmented Planning for Text-Guided Protein Design (2026.findings-acl)

Challenge:	Evidence-intensive reports often produce fluent but under-supported drafts . eviReport is an evidence-grounded workflow for automated long-form report generation .
Approach:	They propose an evidence-tracked workflow that organizes corpus evidence into compact, traceable units and retrieves query-relevant subgraphs into retrieval-ready packages.
Outcome:	The proposed workflow outperforms baselines in factual coverage, factual accuracy and visual evidence integration.

Copied to clipboard

Yutang Ge, Guojiang Zhao, Sihang Li, Zheng Cheng, Zifeng Zhao, Hanchen Xia, Guolin Ke, Linfeng Zhang, Zhifeng Gao, Yu Guang Wang

Challenge:	Recent deep generative models have already shown encouraging * Equal contribution.
Approach:	They propose to use generic instruction-tuned LLMs as direct text-to-sequence generators to achieve this goal.
Outcome:	Recent studies show that reflection improves sequence quality and alignment while maintaining competitive foldability.

OssCSE: Overcoming Surface Structure Bias in Contrastive Learning for Unsupervised Sentence Embedding (2023.emnlp-main)

Copied to clipboard

Zhan Shi, Guoyin Wang, Ke Bai, Jiwei Li, Xiang Li, Qingjun Cui, Belinda Zeng, Trishul Chilimbi, Xiaodan Zhu

SAM Decoding: Speculative Decoding via Suffix Automaton (2025.acl-long)

Challenge:	Recent studies show that contrastive learning is effective in sentence representation learning . but, the surface structure bias is a problem in the current model .
Approach:	They propose to combine a sentence with a sub-semantic sentence to investigate the surface structure bias.
Outcome:	The proposed model achieves state-of-the-art on standard semantic textual similarity tasks using different pre-trained backbones.

Copied to clipboard

Yuxuan Hu, Ke Wang, Xiaokang Zhang, Fanjin Zhang, Cuiping Li, Hong Chen, Jing Zhang

Green & Sustainable NLP

Challenge:	Speculative decoding (SD) methods are inefficient and rely on single retrieval resources.
Approach:	They propose a retrieval-based speculative decoding method that adapts the suffix automaton for efficient draft generation by utilizing the generating text sequence and static text corpus.
Outcome:	The proposed method can find the longest suffix match and can be integrated with existing methods to generalize to broader domains.

Joint Audio/Text Training for Transformer Rescorer of Streaming Speech Recognition (2022.findings-emnlp)

Copied to clipboard

Suyoun Kim, Ke Li, Lucas Kabela, Ron Huang, Jiedan Zhu, Ozlem Kalinli, Duc Le

Bridging the Preference Gap between Retrievers and LLMs (2024.acl-long)

Challenge:	Recent studies have shown that streaming end-to-end speech recognition models suffer from higher word error rates (WER) compared to non-streaming models, streaming endto-ended ASR models are limited to short audio context or not use future context to satisfy low latency constraints.
Approach:	They propose a 2nd-pass rescoring model on top of the 1st-pass streaming model to improve recognition accuracy while keeping latency low.
Outcome:	The proposed method improves word error rate significantly compared to the existing model without adding any additional parameters or latency.

Copied to clipboard

Zixuan Ke, Weize Kong, Cheng Li, Mingyang Zhang, Qiaozhu Mei, Michael Bendersky

EquiBench: Benchmarking Large Language Models’ Reasoning about Program Semantics via Equivalence Checking (2025.emnlp-main)

Challenge:	Existing studies on retrievers and LLMs treat them as separate components . a novel bridge model is proposed to optimize the relationship between the retriever and the LLM .
Approach:	They propose a framework that chains together supervised and reinforcement learning to train a bridge model that optimizes the connection between the retriever and the LLM.
Outcome:	Empirical results show that the proposed model optimizes the connection between the retriever and the LLM.

Copied to clipboard

Anjiang Wei, Jiannan Cao, Ran Li, Hongyu Chen, Yuhui Zhang, Ziheng Wang, Yuan Liu, Thiago S. F. X. Teixeira, Diyi Yang, Ke Wang, Alex Aiken

Open World Classification with Adaptive Negative Samples (2022.emnlp-main)

Challenge:	EquiBench is a new benchmark to evaluate large language models' ability to reason about program semantics . Unlike natural language, code is executable.
Approach:	They propose a benchmark to evaluate large language models through equivalence checking . EquiBench consists of 2400 program pairs across four languages and six categories .
Outcome:	The proposed benchmark consists of 2400 program pairs across four languages and six categories.

Copied to clipboard

Ke Bai, Guoyin Wang, Jiwei Li, Sunghyun Park, Sungjin Lee, Puyang Xu, Ricardo Henao, Lawrence Carin

Hyperlink-induced Pre-training for Passage Retrieval in Open-domain Question Answering (2022.acl-long)

Challenge:	Existing models with no effective open category data during training are limited by the lack of effective open categories data during the training stage.
Approach:	They propose an approach to generate effective open category samples in the training stage and without requiring prior knowledge or external datasets.
Outcome:	The proposed approach generates effective synthetic open category samples in the training stage and without requiring any prior knowledge or external datasets.

Copied to clipboard

Jiawei Zhou, Xiaoguang Li, Lifeng Shang, Lan Luo, Ke Zhan, Enrui Hu, Xinyu Zhang, Hao Jiang, Zhao Cao, Fan Yu, Xin Jiang, Qun Liu, Lei Chen