Papers by Zhiyuan Yang

Repairing Catastrophic-Neglect in Text-to-Image Diffusion Models via Attention-Guided Feature Enhancement (2024.findings-emnlp)

Challenge:	Observable changes in the scene are reflected in captions, but actions are also linked to social aspects such as intentions, effects, and attributes that describe the agent.
Approach:	They propose to generate captions from videos that describe latent aspects of the human agent's actions.
Outcome:	The proposed model can be used to describe latent aspects of human actions in video clips and answer questions about videos.

Copied to clipboard

Zhiyuan Chang, Mingyang Li, Junjie Wang, Yi Liu, Qing Wang, Yang Liu

On LLM-Based Scientific Inductive Reasoning Beyond Equations (2025.emnlp-main)

Challenge:	Text-to-Image Diffusion models generate high-quality images from textual descriptions, but they often produce images that do not fully align with the input prompts, resulting in semantic inconsistencies.
Approach:	They propose an automated repair approach to address catastrophic-neglect in T2I DMs.
Outcome:	The proposed model achieves 10.1%-16.3% higher Correct Rate in image generation compared to baselines.

Copied to clipboard

Brian S. Lin, Jiaxin Yuan, Zihan Zhou, Shouli Wang, Shuo Wang, Cunliang Kong, Qi Shi, Yuxuan Li, Liner Yang, Zhiyuan Liu, Maosong Sun

Long Text and Multi-Table Summarization: Dataset and Method (2022.findings-emnlp)

Challenge:	Existing research on inductive reasoning models emphasizes rule design without grounding them in specific scenarios.
Approach:	They propose to use LLMs to learn underlying patterns from limited examples in entirely new environments.
Outcome:	The proposed benchmark evaluates the inductive reasoning abilities of large language models in scientific settings.

Copied to clipboard

Shuaiqi Liu, Jiannong Cao, Ruosong Yang, Zhiyuan Wen

Enhancing Automated Essay Scoring Performance via Fine-tuning Pre-trained Language Models with Combination of Regression and Ranking (2020.findings-emnlp)

Challenge:	Existing document summarization methods focus on the text and filter out the non-textual content. Existing methods cannot meet the requirements of summarizing long text and multiple tables in each report.
Approach:	They propose a dataset for automatic document summarization that uses text and tabular data to produce a concise summary covering the input document's salient information.
Outcome:	The proposed method can produce a concise summary covering the input document's salient information.

Copied to clipboard

Ruosong Yang, Jiannong Cao, Zhiyuan Wen, Youzheng Wu, Xiaodong He

Seq1F1B: Efficient Sequence-Level Pipeline Parallelism for Large Language Model Training (2025.naacl-long)

Challenge:	Recent work on sentence prediction tasks uses shallow neural networks to learn essay representations and constrain calculated scores with regression loss or ranking loss.
Approach:	They propose to use a pre-trained language model to learn text representations first and then to constrain the scores with regression loss or ranking loss.
Outcome:	The proposed model outperforms state-of-the-art models on the Automated Student Assessment Prize dataset.

Copied to clipboard

Sun Ao, Weilin Zhao, Xu Han, Cheng Yang, Xinrong Zhang, Zhiyuan Liu, Chuan Shi, Maosong Sun

More Data, More Relations, More Context and More Openness: A Review and Outlook for Relation Extraction (2020.aacl-main)

Challenge:	Current PP methods face severe bottlenecks, including pipeline bubbles and memory footprint.
Approach:	They propose a sequence-level one-forward-one-backward (1F1B) PP method for training LLMs on long sequences with high throughput and memory efficiency.
Outcome:	The proposed method achieves 1.14X training throughput with half memory footprint compared to baseline methods . it trains an LLM with 30B parameters on sequences up to 64k tokens using 64X NVIDIA A100 GPUs .

Copied to clipboard

Xu Han, Tianyu Gao, Yankai Lin, Hao Peng, Yaoliang Yang, Chaojun Xiao, Zhiyuan Liu, Peng Li, Jie Zhou, Maosong Sun

Information Extraction & Text Mining

Challenge:	Existing methods for extracting relational facts from text have been successful . but with explosion of Web text, human knowledge is increasing drastically .
Approach:	They propose to improve relation extraction methods to extract relational facts from text . they analyze existing methods and show promising directions towards more powerful RE .
Outcome:	The proposed methods can extract relational facts from text, but they are still lacking in the current field.

KACC: A Multi-task Benchmark for Knowledge Abstraction, Concretization and Completion (2021.findings-acl)

Copied to clipboard

Jie Zhou, Shengding Hu, Xin Lv, Cheng Yang, Zhiyuan Liu, Wei Xu, Jie Jiang, Juanzi Li, Maosong Sun

Optima: Optimizing Effectiveness and Efficiency for LLM-Based Multi-Agent System (2025.findings-acl)

Challenge:	Existing studies focus on partial aspects of knowledge abstraction, concretization, and completion (KACC).
Approach:	They propose a unified knowledge graph benchmark to improve existing benchmarks . they collect new datasets that contain larger concept graphs and cross-view links .
Outcome:	The proposed benchmark improves existing benchmarks in terms of dataset scale, task coverage, and difficulty.

Copied to clipboard

Weize Chen, Jiarui Yuan, Chen Qian, Cheng Yang, Zhiyuan Liu, Maosong Sun

Green & Sustainable NLP

Challenge:	Large Language Models (LLMs) have emerged as powerful tools for a wide range of tasks, from * Equal Contribution.
Approach:	They propose a framework that enhances communication efficiency and task effectiveness in LLM-based multi-agent systems through training.
Outcome:	The proposed framework improves communication efficiency and task effectiveness on multi-agent tasks with 2.8x performance gain with less than 10% tokens on tasks requiring heavy information exchange.

AgentCPM-GUI: Building Mobile-Use Agents with Reinforcement Fine-Tuning (2025.emnlp-demos)

Copied to clipboard

Zhong Zhang, Yaxi Lu, Yikun Fu, Yupeng Huo, Shenzhi Yang, Yesai Wu, Han Si, Xin Cong, Haotian Chen, Yankai Lin, Xie Xie, Wei Zhou, Wang Xu, Zhou Su, Zhongwu Zhai, Xiaoming Liu, null Meiyudong, Jianming Xu, Hongyan Tian, Chongyi Wang, Chi Chen, Yuan Yao, Zhiyuan Liu, Maosong Sun

Judge as A Judge: Improving the Evaluation of Retrieval-Augmented Generation through the Judge-Consistency of Large Language Models (2025.findings-acl)

Challenge:	Large language model agents have enabled GUI-based automation, but their deployment is limited by noisy data, poor generalization, and lack of support for non-English GUIs.
Approach:	They propose an 8B-parameter GUI agent built for robust and efficient on-device GUI interaction.
Outcome:	The proposed GUI agent achieves promising performance on five public benchmarks and proposed Chinese benchmark CAGUI.

Copied to clipboard

Shuliang Liu, Xinze Li, Zhenghao Liu, Yukun Yan, Cheng Yang, Zheni Zeng, Zhiyuan Liu, Maosong Sun, Ge Yu

GEAR: Graph-based Evidence Aggregating and Reasoning for Fact Verification (P19-1)

Challenge:	Existing evaluation metrics cannot fairly evaluate the outputs of RAG models during training and evaluation.
Approach:	They propose a method which prompts LLMs to generate different judgments based on various combinations of judgment dimensions and utilizes the judge-consistency to evaluate these judgments.
Outcome:	The proposed method generates more accurate evaluations for RAG models across different RAG model and datasets.

Copied to clipboard

Jie Zhou, Xu Han, Cheng Yang, Zhiyuan Liu, Lifeng Wang, Changcheng Li, Maosong Sun

MolTC: Towards Molecular Relational Modeling In Language Models (2024.findings-acl)

Challenge:	Existing methods to extract information from evidence are unable to grasp relational and logical information among the evidence.
Approach:	They propose a graph-based evidence aggregating and reasoning framework to integrate evidence from multiple pieces of evidence.
Outcome:	The proposed framework achieves significant performance improvements on a large-scale benchmark dataset.

Copied to clipboard

Junfeng Fang, Shuai Zhang, Chang Wu, Zhengyi Yang, Zhiyuan Liu, Sihang Li, Kun Wang, Wenjie Du, Xiang Wang

RepoGenesis: Benchmarking End-to-End Microservice Generation from Readme to Repository (2026.acl-long)

Challenge:	Molecular Relational Learning (MRL) is a promising way to understand interactions between molecular pairs.
Approach:	They propose a novel LLM-based multi-modal framework for molecular interaction modeling following Chain-of-Thought (CoT) theory which integrates graphical information of two molecules in pair.
Outcome:	The proposed framework integrates graphical information of two molecules in pair.

Copied to clipboard

Zhiyuan Peng, Xin Yin, Pu Zhao, Fangkai Yang, Lu Wang, Ran Jia, Xu Chen, Qingwei Lin, Saravan Rajmohan, Dongmei Zhang

Challenge:	Existing benchmarks focus on isolated function/class-level generation, neglecting complete microservice repository generation.
Approach:	They propose a multilingual benchmark for repository-level end-to-end web microservice generation that reflects real-world development workflows.
Outcome:	The benchmark compared 106 repositories across 18 domains and 11 frameworks and 1,258 API endpoints and 2,335 test cases.

Word-level Textual Adversarial Attacking as Combinatorial Optimization (2020.acl-main)

Copied to clipboard

Yuan Zang, Fanchao Qi, Chenghao Yang, Zhiyuan Liu, Meng Zhang, Qun Liu, Maosong Sun

Robustness in NLP

Challenge:	Existing word-level attack models are far from perfect because of unsuitable search space reduction methods and inefficient optimization algorithms.
Approach:	They propose a novel adversarial adversarialist model that incorporates word substitution and particle swarm optimization to solve two problems separately.
Outcome:	The proposed model achieves much higher success rates and crafts more high-quality adversarial examples as compared to baseline methods.

Decode with Template: Content Preserving Sentiment Transfer (2020.lrec-1)

Copied to clipboard

Zhiyuan Wen, Jiannong Cao, Ruosong Yang, Senzhang Wang

Sentiment Analysis

Challenge:	Existing methods to transfer sentiments for text use only explicit sentiments and templates to remove them from input sentences.
Approach:	They propose a method to transfer sentiments from input sentences to output sentences using templates.
Outcome:	The proposed model significantly outperforms state-of-the-art models in content preservation.

Automatically Select Emotion for Response via Personality-affected Emotion Transition (2021.findings-acl)

Copied to clipboard

Zhiyuan Wen, Jiannong Cao, Ruosong Yang, Shuaiqi Liu, Jiaxing Shen

Dialogue Systems & Conversational Agents

Challenge:	Existing studies focus on rendering specified emotions in responses, yet the individual difference in emotion expression is overlooked.
Approach:	They propose to equip a dialog system with personality and enable it to select emotions in responses like humans.
Outcome:	The proposed system can select emotions in responses like humans by simulating the emotion transition of humans in conversation.

Enhancing Long-Chain Reasoning Distillation through Error-Aware Self-Reflection (2026.findings-acl)

Copied to clipboard

Zhuoyang Wu, Xinze Li, Zhenghao Liu, Yukun Yan, Zhiyuan Liu, Minghe Yu, Cheng Yang, Yu Gu, Ge Yu, Maosong Sun

Beyond Natural Language: LLMs Leveraging Alternative Formats for Enhanced Reasoning and Communication (2024.findings-emnlp)

Challenge:	Existing studies treat SLMs as student models and use long-form Chains-of-Thought (CoTs) as supervision signals for Supervised Fine-Tuning (SFT). Existing research focuses on distilling reasoning ability from LLMs to enhance the mathematical reasoning performance of small-scale models.
Approach:	They propose a framework that refines teacher CoTs through an error-aware reflection process to enable the student model to construct more tailored teacher Cots.
Outcome:	Experiments on multiple mathematical reasoning benchmarks show that ORION improves performance by more than 2% over all baselines.

Copied to clipboard

Weize Chen, Chenfei Yuan, Jiarui Yuan, Yusheng Su, Chen Qian, Cheng Yang, Ruobing Xie, Zhiyuan Liu, Maosong Sun

End-to-end Knowledge Retrieval with Multi-modal Queries (2023.acl-long)

Challenge:	Natural language (NL) has long been the predominant format for human cognition and communication, but its utility in LLMs has not been thoroughly examined.
Approach:	They propose to allow LLMs to choose the most suitable format before reasoning or communicating, and to automate the selection process.
Outcome:	The proposed format improves reasoning efficiency and reduces token usage while maintaining communicative effectiveness.

Copied to clipboard

Man Luo, Zhiyuan Fang, Tejas Gokhale, Yezhou Yang, Chitta Baral

Play Guessing Game with LLM: Indirect Jailbreak Attack with Implicit Clues (2024.findings-acl)

Challenge:	a new task is proposed to learn knowledge retrieval with multimodal queries . a vision-language model can retrieve knowledge using images and text inputs .
Approach:	They propose a task for vision-language models to retrieve knowledge with multi-modal queries . they propose reViz, a model that integrates content from both text and image queries based on a multimodal query task .
Outcome:	The proposed task performs better under zero-shot settings than previous work on cross-modal retrieval.

Copied to clipboard

Zhiyuan Chang, Mingyang Li, Yi Liu, Junjie Wang, Qing Wang, Yang Liu

CAR: Empowering Agents with Dynamic Tool Synthesis and Global Trajectory Rectification (2026.findings-acl)

Challenge:	Existing jailbreak attacks primarily utilize scenario camouflage techniques, however their explicit mention of malicious intent will be easily recognized and defended by LLMs.
Approach:	They propose an indirect jailbreak attack approach, Puzzler, which can bypass LLM’s defensive strategies and obtain malicious response by implicitly providing LLMs with some clues about the original malicious query.
Outcome:	The proposed approach can bypass the LLM’s defensive strategies and obtain malicious response by implicitly providing LLMs with some clues about the original malicious query.

Copied to clipboard

Kai Wang, Xu Wang, Zhiyuan Fu, Yudong Zhang, Yang Wang

Challenge:	Existing LLM agents are brittle in open-ended environments due to two limitations: 1) a closed action space; 2) myopic error recovery.
Approach:	They propose a novel architecture that augments the action space and revises global strategies by adding a reflective replanning mechanism to the system.
Outcome:	Experiments show that CAR outperforms baselines in a diagnostic benchmark with pruned toolsets to simulate tool scarcity.

Be Careful about Poisoned Word Embeddings: Exploring the Vulnerability of the Embedding Layers in NLP Models (2021.naacl-main)

Copied to clipboard

Wenkai Yang, Lei Li, Zhiyuan Zhang, Xuancheng Ren, Xu Sun, Bin He

Representation Learning

Challenge:	Recent studies reveal a security threat to natural language processing models, called the Backdoor Attack.
Approach:	They propose to hack a model by modifying one single word embedding vector without sacrificing accuracy on clean samples.
Outcome:	The proposed method is more efficient and stealthier on sentiment analysis and sentence-pair classification tasks.

MatPlotAgent: Method and Evaluation for LLM-Based Agentic Scientific Data Visualization (2024.findings-acl)

Copied to clipboard

Zhiyu Yang, Zihan Zhou, Shuo Wang, Xin Cong, Xu Han, Yukun Yan, Zhenghao Liu, Zhixing Tan, Pengyuan Liu, Dong Yu, Zhiyuan Liu, Xiaodong Shi, Maosong Sun

Exploring the Impact of Model Scaling on Parameter-Efficient Tuning (2023.emnlp-main)

Challenge:	Scientific data visualization is an essential process in research, but its use of large language models remains unexplored.
Approach:	They propose a model-agnostic LLM agent framework to automate scientific data visualization tasks.
Outcome:	The proposed framework improves performance of commercial and open-source models.

Copied to clipboard

Yusheng Su, Chi-Min Chan, Jiali Cheng, Yujia Qin, Yankai Lin, Shengding Hu, Zonghan Yang, Ning Ding, Xingzhi Sun, Guotong Xie, Zhiyuan Liu, Maosong Sun

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

Challenge:	Parameter-efficient tuning (PET) methods can drive large pre-trained language models by training only minimal parameters.
Approach:	They propose a parameter-efficient tuning method that is compatible with a tunable module and uses a random number generator to optimize fewer table parameters.
Outcome:	The proposed method is compatible with a tunable module and tested on 11 NLP tasks.

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

CodexGraph: Bridging Large Language Models and Code Repositories via Code Graph Databases (2025.naacl-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

Xiangyan Liu, Bo Lan, Zhiyuan Hu, Yang Liu, Zhicheng Zhang, Fei Wang, Michael Qizhe Shieh, Wenmeng Zhou

Mimicking the Familiar: Dynamic Command Generation for Information Theft Attacks in LLM Tool-Learning System (2025.acl-long)

Challenge:	Large Language Models excel in stand-alone code tasks but struggle with handling entire code repositories.
Approach:	They propose a system that integrates LLM agents with graph database interfaces extracted from code repositories.
Outcome:	The proposed system integrates LLM agents with graph database interfaces extracted from code repositories.

Copied to clipboard

Ziyou Jiang, Mingyang Li, Guowei Yang, Junjie Wang, Yuekai Huang, Zhiyuan Chang, Qing Wang

Robustness in NLP

Challenge:	Existing approaches to attack Large Language Model (LLM) tool-learning systems are black-box oriented and rely on static commands that cannot adapt flexibly to the changes in user queries and the invocation chain.
Approach:	They propose a dynamic attack comment generation approach for information theft attacks in LLM tool-learning systems that mimics the familiar by inferring the information utilized by upstream tools.
Outcome:	The proposed approach outperforms baselines with +13.2% ASRTheft and can be generalized to new tool-learning systems to expose their information leakage risks.

LeanK: Learnable K Cache Channel Pruning for Efficient Decoding (2025.emnlp-main)

Copied to clipboard

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

Green & Sustainable NLP

Challenge:	Existing efforts to optimize the key-value (KV) cache include: (1) Eviction, which discards cache of less important tokens; (2) Selection, which retains the full KV cache but selectively reads relevant entries.
Approach:	They propose a learning-based method that prunes unimportant key (K) cache channels by leveraging static channel sparsity.
Outcome:	Experiments show that LeanK reduces GPU memory and accelerates decoding without sacrificing accuracy.

Robust and Scalable Model Editing for Large Language Models (2024.lrec-main)

Copied to clipboard

Yingfa Chen, Zhengyan Zhang, Xu Han, Chaojun Xiao, Zhiyuan Liu, Chen Chen, Kuai Li, Tao Yang, Maosong Sun

A Query-Response Framework for Whole-Page Complex-Layout Document Image Translation with Relevant Regional Concentration (2025.findings-acl)

Challenge:	Existing methods that ignore contextual knowledge fail to reliably fall back to parametric knowledge when presented with irrelevant context.
Approach:	They propose to use contextual knowledge to update and correct LLMs' knowledge by in-context editing instead of retraining.
Outcome:	The proposed method outperforms current state-of-the-art methods by a large margin on a dataset that contains irrelevant questions.

Copied to clipboard

Zhiyang Zhang, Yaping Zhang, Yupu Liang, Zhiyuan Chen, Lu Xiang, Yang Zhao, Yu Zhou, Chengqing Zong

Machine Translation

Challenge:	Existing methods for document image translation rely on the vanilla encoder-decoder paradigm . a novel dynamic aggregation mechanism is designed to enhance the text semantics in query features toward translation.
Approach:	They propose a Query-Response DIT framework that reformulates the DIT task into a parallel response/translation process of multiple queries.
Outcome:	The proposed framework improves translation quality on four translation directions on three benchmarks.

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

Copied to clipboard

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

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

QuoteR: A Benchmark of Quote Recommendation for Writing (2022.acl-long)

Copied to clipboard

Fanchao Qi, Yanhui Yang, Jing Yi, Zhili Cheng, Zhiyuan Liu, Maosong Sun

Challenge:	Existing methods to recommend quotes are evaluated on unpublished datasets .
Approach:	They propose to build a dataset that is open and contains three parts including English, standard Chinese and classical Chinese.
Outcome:	The proposed model outperforms existing methods on all three parts of QuoteR.

Position Engineering: Boosting Large Language Models through Positional Information Manipulation (2024.emnlp-main)

Copied to clipboard

Zhiyuan He, Huiqiang Jiang, Zilong Wang, Yuqing Yang, Luna Qiu, Lili Qiu

StableToolBench: Towards Stable Large-Scale Benchmarking on Tool Learning of Large Language Models (2024.findings-acl)

Challenge:	Recent advances in Large Language Models (LLMs) have demonstrated significant strides towards achieving artificial general intelligence.
Approach:	They propose a technique termed position engineering which alters the positional information in the prompt without modifying the text itself.
Outcome:	The proposed technique significantly improves on the baseline in retrieval-augmented generation and in-context learning scenarios.

Copied to clipboard

Zhicheng Guo, Sijie Cheng, Hao Wang, Shihao Liang, Yujia Qin, Peng Li, Zhiyuan Liu, Maosong Sun, Yang Liu

A Language Model based Evaluator for Sentence Compression (P18-2)

Challenge:	Large Language Models (LLMs) have witnessed remarkable advancements in recent years, prompting the exploration of tool learning.
Approach:	They propose a virtual API server and stable evaluation system to assess the stability of large-scale real-time APIs.
Outcome:	The proposed benchmarks demonstrate the stability of the proposed system and its caching system.

Copied to clipboard

Yang Zhao, Zhiyuan Luo, Akiko Aizawa

A Survey of Retentive Network (2026.findings-acl)

Challenge:	Existing methods to produce readable sentence compression are based on machine learning or syntactic tree-based approaches.
Approach:	They propose a language-model-based evaluator for deletion-based sentence compression . they propose deleting operations on source sentences to obtain best target compression based on the proposed model .
Outcome:	The proposed model generates more readable compression comparable to strong baselines.

Copied to clipboard

Haiqi Yang, Zhiyuan Li, Yi Chang, Yuan Wu

Challenge:	Existing studies on the effectiveness of the Retentive Networks have not yet been conducted.
Approach:	They propose a retention mechanism that integrates the inductive bias of recurrent neural networks with the parallelizable training advantages of attention-based models.
Outcome:	The proposed retention mechanism combines the inductive bias of recurrent neural networks with the parallelizable training advantages of attention-based models.

Expose Backdoors on the Way: A Feature-Based Efficient Defense against Textual Backdoor Attacks (2022.findings-emnlp)

Copied to clipboard

Sishuo Chen, Wenkai Yang, Zhiyuan Zhang, Xiaohan Bi, Xu Sun

Experiential Co-Learning of Software-Developing Agents (2024.acl-long)

Challenge:	Existing online backdoor defense methods for NLP models focus on anomalies at input or output level, causing fragility to adaptive attacks and high computational cost.
Approach:	They propose a feature-based online defense method to detect poisoned samples . they use a distance-based anomaly score to distinguish poisones from clean samples based on feature-level regularization .
Outcome:	The proposed method outperforms existing methods in sentiment analysis and offense detection tasks.

Copied to clipboard

Chen Qian, Yufan Dang, Jiahao Li, Wei Liu, Zihao Xie, YiFei Wang, Weize Chen, Cheng Yang, Xin Cong, Xiaoyin Che, Zhiyuan Liu, Maosong Sun

One Shot Dominance: Knowledge Poisoning Attack on Retrieval-Augmented Generation Systems (2025.findings-emnlp)

Challenge:	Recent advances in large language models (LLMs) have brought significant changes to various domains, especially through autonomous agents.
Approach:	They propose a framework that lets agents learn shortcuts from their past tasks and use them for future task execution.
Outcome:	The proposed framework enables agents to tackle unseen software-developing tasks more effectively.

Copied to clipboard

Zhiyuan Chang, Mingyang Li, Xiaojun Jia, Junjie Wang, Yuekai Huang, Ziyou Jiang, Yang Liu, Qing Wang

Robustness in NLP

Challenge:	Existing knowledge poisoning attacks against RAG systems require multiple poisoned documents or can only function effectively on simplistic queries.
Approach:	They propose a more realistic knowledge poisoning attack that poisons only a single document while remaining effective for complex multi-hop questions involving complex relationships between multiple elements.
Outcome:	The proposed attack achieves success by poisoning only a single document while remaining effective for complex multi-hop questions involving complex relationships between multiple elements.

Progressive Self-Training with Discriminator for Aspect Term Extraction (2021.emnlp-main)

Copied to clipboard

Qianlong Wang, Zhiyuan Wen, Qin Zhao, Min Yang, Ruifeng Xu

Information Extraction & Text Mining

Challenge:	Existing approaches to extract aspect terms from review sentences are limited due to lack of annotated data.
Approach:	They propose to refine conventional self-training to progressive self-teaching to reduce noise . they use a discriminator to filter the noisy pseudo-labels.
Outcome:	The proposed model outperforms baseline models and achieves state-of-the-art performance on four SemEval datasets.

TSPO: Breaking the Double Homogenization Dilemma in Multi-turn Search Policy Optimization (2026.findings-acl)

Copied to clipboard

Shichao Ma, Zhiyuan Ma, Ming Yang, Xiaofan Li, Xing Wu, Jintao Du, Yu Cheng, Weiqiang Wang, Qiliang Liu, Zhengyang Zhou, Yang Wang

Improving MLLM’s Document Image Machine Translation via Synchronously Self-reviewing Its OCR Proficiency (2025.findings-acl)

Challenge:	Large Language Models (LLMs) can solve complex tasks through iterative information retrieval.
Approach:	They propose a turn-level stage-aware policy optimization approach to solve this problem . they introduce a first-occurrence latent reward mechanism to allocate partial rewards .
Outcome:	Experiments show that TSPO outperforms state-of-the-art models on Qwen2.5-3B and 7B models.

Copied to clipboard

Yupu Liang, Yaping Zhang, Zhiyang Zhang, Zhiyuan Chen, Yang Zhao, Lu Xiang, Chengqing Zong, Yu Zhou

Multilinguality

Challenge:	Multimodal Large Language Models (MLLMs) have shown strong performance in document image tasks, especially Optical Character Recognition (OCR). However, they struggle with Document Image Machine Translation (DIMT), which requires handling both cross-modal and cross-lingual challenges.
Approach:	They propose a novel fine-tuning paradigm that allows the model to generate OCR text before producing translation text, which allows it to leverage its strong monolingual OCR ability while learning to translate text across languages.
Outcome:	The proposed model can leverage its strong monolingual OCR ability while learning to translate text across languages.

WantWords: An Open-source Online Reverse Dictionary System (2020.emnlp-demos)

Copied to clipboard

Fanchao Qi, Lei Zhang, Yanhui Yang, Zhiyuan Liu, Maosong Sun

Multilinguality

Challenge:	Existing reverse dictionary systems only support English reverse dictionary queries . a reverse dictionary can help people who can't remember a word from memory .
Approach:	They propose an online reverse dictionary system that outperforms other reverse dictionary systems . it supports Chinese and English-Chinese as well as Chinese-English cross-lingual reverse dictionary queries .
Outcome:	The proposed reverse dictionary outperforms other reverse dictionary systems on performance . it supports Chinese and English-Chinese as well as Chinese-English queries .

UltraLink: An Open-Source Knowledge-Enhanced Multilingual Supervised Fine-tuning Dataset (2024.acl-long)

Copied to clipboard

Haoyu Wang, Shuo Wang, Yukun Yan, Xujia Wang, Zhiyu Yang, Yuzhuang Xu, Zhenghao Liu, Liner Yang, Ning Ding, Xu Han, Zhiyuan Liu, Maosong Sun

RealMem: Benchmarking LLMs in Real-World Memory-Driven Interaction (2026.findings-acl)

Challenge:	Open-source large language models (LLMs) have gained strength across diverse fields, but the majority of studies focus on English.
Approach:	They propose a knowledge-grounded data augmentation approach to elicit more language-specific knowledge of LLMs by enhancing their ability to serve users from different countries.
Outcome:	The proposed method can prune the language-agnostic supervised fine-tuning dataset without any performance degradation.

Copied to clipboard

Haonan Bian, Zhiyuan Yao, Sen Hu, Zishan Xu, Shaolei Zhang, Yifu Guo, Ziliang Yang, Xueran Han, Huacan Wang, Ronghao Chen

Dialogue Systems & Conversational Agents

Challenge:	Existing benchmarks focus on casual conversation or task-oriented dialogue, failing to capture “long-term project-oriented” interactions where agents must track evolving goals.
Approach:	They propose a benchmark that simulates the dynamic evolution of memory in real-world projects.
Outcome:	The proposed benchmarks simulate the dynamic evolution of memory in real-world projects.

COAST: Enhancing the Code Debugging Ability of LLMs through Communicative Agent Based Data Synthesis (2025.findings-naacl)

Copied to clipboard

Weiqing Yang, Hanbin Wang, Zhenghao Liu, Xinze Li, Yukun Yan, Shuo Wang, Yu Gu, Minghe Yu, Zhiyuan Liu, Ge Yu

Large Language Models Are Still Misled by Simple Bias Ensembles (2026.findings-acl)

Challenge:	Existing code debugging benchmarks focus on the Code Repair stage of the code generation process.
Approach:	They propose a framework to evaluate the debugging abilities of large language models by emulating the human debug process.
Outcome:	The proposed framework outperforms human-curated and GPT-4-generated training data, enabling 7B-scale LLMs to achieve comparable debugging performance to GPT-3.5.

Copied to clipboard

Zhouhao Sun, Zhiyuan Kan, Xiao Ding, Li Du, Bibo Cai, Yang Zhao, Bing Qin, Ting Liu

Highlight-Transformer: Leveraging Key Phrase Aware Attention to Improve Abstractive Multi-Document Summarization (2021.findings-acl)

Challenge:	Existing benchmarks for large language models are constrained to datasets where each sample is manually injected with only one type of bias.
Approach:	They propose a multi-bias benchmark where each sample contains multiple types of biases.
Outcome:	The proposed benchmark shows that existing LLMs and debiasing methods perform poorly on this benchmark, highlighting the challenge of eliminating compounded biases.

Copied to clipboard

Shuaiqi Liu, Jiannong Cao, Ruosong Yang, Zhiyuan Wen

Unleashing Low-Bit Inference on Ascend NPUs: A Comprehensive Evaluation of HiFloat Formats (2026.acl-industry)

Challenge:	Existing models do not consider key phrases in determining attention weights of self-attention . Existing work does not consider the importance of key phrases when determining weights .
Approach:	They propose a model with highlighting mechanism to assign greater attention weights to key phrases . they propose two structures of highlighting attention for each head and the multihead highlighting . experimental results show that their proposed model significantly outperforms the baseline model .
Outcome:	The proposed model outperforms the baseline models on a multi-news dataset.

Copied to clipboard

Pengxiang Zhao, Hui-Ling Zhen, Xing Li, Han Bao, Weizhe Lin, Zhiyuan Yang, Yu Zi Wei, Xin Wang, Mingxuan Yuan, Xianzhi Yu, Zhenhua Dong

Green & Sustainable NLP

Challenge:	Low-bit floating-point formats like MXFP and NVFP4 offer new opportunities for precision and efficiency.
Approach:	They evaluate HiFloat (HiF8 and HiF4), a family of floating-point formats tailored for Ascend NPUs.
Outcome:	The proposed formats excel with high-variance data and are compatible with state-of-the-art quantization frameworks.

Modeling Semantic Compositionality with Sememe Knowledge (P19-1)

Copied to clipboard

Fanchao Qi, Junjie Huang, Chenghao Yang, Zhiyuan Liu, Xiao Chen, Qun Liu, Maosong Sun

Enhancing Open-Domain Task-Solving Capability of LLMs via Autonomous Tool Integration from GitHub (2025.acl-long)

Challenge:	Semantic compositionality (SC) is defined as the phenomenon that the meaning of a complex linguistic unit can be composed of the meanings of its constituents.
Approach:	They propose to incorporate sememes into SC models and employ them in learning multiword expressions.
Outcome:	The proposed models achieve significant performance boost compared to baseline methods without sememe knowledge.

Copied to clipboard

Bohan Lyu, Xin Cong, Heyang Yu, Pan Yang, Cheng Qian, Zihe Wang, Yujia Qin, Yining Ye, Yaxi Lu, Chen Qian, Zhong Zhang, Yukun Yan, Yankai Lin, Zhiyuan Liu, Maosong Sun

ACE-Router: Generalizing History-Aware Routing from MCP Tools to the Agent Web (2026.acl-long)

Challenge:	Existing approaches lack flexibility to address diverse and ever-evolving user queries in open domains.
Approach:	They propose to evaluate LLMs on open-domain knowledge that requires tools to solve diverse and ever-evolving user queries.
Outcome:	The proposed system outperforms baselines in the open domain task-solving benchmark.

Copied to clipboard

Zhiyuan Yao, Zishan Xu, Yifu Guo, Zhiguang Han, Cheng Yang, Shuo Zhang, Weinan Zhang, Xingshan Zeng, Weiwen Liu

Challenge:	Existing routers that use hardcoded tools are limited by scalability and generality bottlenecks.
Approach:	They propose a pipeline for training history-aware routers to empower precise navigation in large-scale ecosystems.
Outcome:	The proposed pipeline can train routers with dynamic context understanding to create the plug-and-play Light Routing Agent.

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

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