Papers by Fei Yin
HACHIMI: Scalable and Controllable Student Persona Generation via Orchestrated Agents (2026.findings-acl)
Copied to clipboard
| Challenge: | ad-hoc prompting and hand-crafted profiles with limited control over educational theory and population distributions are often used for student personas. |
| Approach: | They propose a framework that generates theory-aligned, quota-controlled personas . they factorize each persona into a theory-anchored educational schema . |
| Outcome: | HACHIMI generates theory-aligned, quota-controlled personas for grades 1-12 . results show near-perfect schema validity, accurate quots, and substantial diversity . |
Too Long, Do Re-weighting for Efficient LLM Reasoning Compression (2026.acl-long)
Copied to clipboard
Zhong-Zhi Li, Xiao Liang, Zihao Tang, Lei Ji, Peijie Wang, Haotian Xu, Xing W, Haizhen Huang, Weiwei Deng, Yeyun Gong, Zhijiang Guo, Xiao Liu, Fei Yin, Cheng-Lin Liu
| Challenge: | Large Language Models (LLMs) have recently achieved remarkable progress on complex reasoning tasks by leveraging extended Chain-of-Thought (CoT) techniques. |
| Approach: | They propose a method that uses Extended Chain-of-Thought (EFT) to reduce the number of output tokens by nearly 40% while maintaining the accuracy of the reasoning. |
| Outcome: | The proposed method reduces the number of output tokens by nearly 40% while maintaining the accuracy of the reasoning. |
LLMs Assist NLP Researchers: Critique Paper (Meta-)Reviewing (2024.emnlp-main)
Copied to clipboard
Jiangshu Du, Yibo Wang, Wenting Zhao, Zhongfen Deng, Shuaiqi Liu, Renze Lou, Henry Zou, Pranav Narayanan Venkit, Nan Zhang, Mukund Srinath, Haoran Zhang, Vipul Gupta, Yinghui Li, Tao Li, Fei Wang, Qin Liu, Tianlin Liu, Pengzhi Gao, Congying Xia, Chen Xing, Cheng Jiayang, Zhaowei Wang, Ying Su, Raj Shah, Ruohao Guo, Jing Gu, Haoran Li, Kangda Wei, Zihao Wang, Lu Cheng, Surangika Ranathunga, Meng Fang, Jie Fu, Fei Liu, Ruihong Huang, Eduardo Blanco, Yixin Cao, Rui Zhang, Philip Yu, Wenpeng Yin
| Challenge: | a comparative analysis of paper (meta-)reviews by large language models (LLMs) aims to identify and distinguish LLMs from human activities . |
| Approach: | They present a comparative analysis to identify and distinguish LLM activities from human activities. |
| Outcome: | The proposed analysis aims to improve recognition of instances when someone implicitly uses LLMs for reviewing activities. |
Beyond Static Testbeds: An Interaction-Centric Agent Simulation Platform for Dynamic Recommender Systems (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing platforms lack a mechanism for user actions to dynamically reshape the environment. |
| Approach: | They propose a novel agent-based simulation platform for recommender systems with a robust interaction mechanism. |
| Outcome: | The proposed platform improves the credibility of the simulation and replicates the Matthew Effect and Brand Loyalty. |
The Mirage of Model Editing: Revisiting Evaluation in the Wild (2025.acl-long)
Copied to clipboard
| Challenge: | despite near-perfect results, effectiveness of model editing in real-world applications remains unclear. |
| Approach: | They propose QAEdit and WILD to better reflect real-world use of model editing . they propose a benchmark aligned with widely used question answering datasets and a task-agnostic evaluation framework . |
| Outcome: | The proposed QAEdit benchmark and WILD evaluation framework show that current models perform worse than previously reported. |
Towards General Agentic Intelligence via Environment Scaling (2026.findings-acl)
Copied to clipboard
Runnan Fang, Shihao Cai, Baixuan Li, Jialong Wu, Guangyu Li, Wenbiao Yin, Xinyu Wang, Xiaobin Wang, Liangcai Su, Zhen Zhang, Shibin Wu, Zhengwei Tao, Yong Jiang, Pengjun Xie, Ningyu Zhang, Fei Huang, Wentao Zhang, Jingren Zhou
| Challenge: | Diverse real-world APIs require precise, robust function-calling intelligence, which needs agents to develop these capabilities through interaction in varied environments. |
| Approach: | They propose a framework that scales up environments to enable agentic intelligence . they use a two-phase agent fine-tuning strategy to first endow agents with basic agentic capabilities, then specializing them for domain-specific contexts. |
| Outcome: | Experiments on -bench, -Bench, and ACEBench show that the model significantly enhances the models’ function-calling capability. |
WESR: A Benchmark and Strong Baseline for Word-level Event-Speech Recognition (2026.findings-acl)
Copied to clipboard
Chenchen Yang, Kexin Huang, Liwei Fan, Qian Tu, Botian Jiang, Dong Zhang, Linqi Yin, Shimin Li, Zhaoye Fei, Qinyuan Cheng, Xipeng Qiu
| Challenge: | aaron carroll: the precise localization of non-verbal vocal events remains a critical yet under-explored challenge. carroll says current methods suffer from insufficient task definitions with limited category coverage. carrol: knowing exactly where an event occurred is not enough; knowing exactly what it happened is. |
| Approach: | They propose a taxonomy of 21 vocal events with a new categorization into discrete versus continuous types. |
| Outcome: | The proposed model disentangles ASR errors from event detection while maintaining ASR quality. |
Turn Waste into Worth: Rectifying Top-k Router of MoE (2024.emnlp-main)
Copied to clipboard
Zhiyuan Zeng, Qipeng Guo, Zhaoye Fei, Zhangyue Yin, Yunhua Zhou, Linyang Li, Tianxiang Sun, Hang Yan, Dahua Lin, Xipeng Qiu
| Challenge: | Top-k router suffers from redundancy computation and memory costs due to unbalanced routing . some experts are overflow, where exceeding tokens are dropped, while others are empty, which are padded with zeros, negatively impacting model performance. |
| Approach: | They propose a top-k router that is unbalanced and uses a multi-gPU system to handle dropped tokens and padding. |
| Outcome: | The proposed model surpasses the top-1 router by 4.7% in terms of performance . the top-k router suffers from redundancy computation and memory costs . |
Focusing Condition: Inference-Time Self-Contrastive Steering Elicits Better Conditional Text Embeddings in LLMs (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods for extracting conditional text embeddings from large language models (LLMs) relying on prompts often fails to produce high-quality conditional embeddables, resulting in degradation of quality. |
| Approach: | They propose a plug-and-play method that constructs unconditional general text embeddings and uses them to refine conditional text embeds. |
| Outcome: | The proposed method improves performance of prompt-based methods on clustering, Semantic Textual Similarity, and triplet alignment datasets. |
One Cannot Stand for Everyone! Leveraging Multiple User Simulators to train Task-oriented Dialogue Systems (2023.acl-long)
Copied to clipboard
| Challenge: | Recent studies have found that Task-oriented Dialogue systems can be more suitable for human users. |
| Approach: | They propose a framework to optimize ToD systems by leveraging Multiple User SimulaTors. |
| Outcome: | The proposed framework improves performance on multiWOZ with human evaluations and automatic evaluations. |
PRDetect: Perturbation-Robust LLM-generated Text Detection Based on Syntax Tree (2025.findings-naacl)
Copied to clipboard
| Challenge: | Recent methods for detecting LLM-generated text have shown impressive performance, but in real-world scenarios, users often introduce perturbations to the text. |
| Approach: | They propose a method that detects syntactic trees that are minimally affected by perturbations and exhibit distinct differences between human-written and LLM-generated text. |
| Outcome: | The proposed method shows that it is significantly better against perturbations on the HC3 and GPT-3.5-mixed datasets and also has the shortest time expenditure. |
OmniThink: Expanding Knowledge Boundaries in Machine Writing through Thinking (2025.emnlp-main)
Copied to clipboard
Zekun Xi, Wenbiao Yin, Jizhan Fang, Jialong Wu, Runnan Fang, Yong Jiang, Pengjun Xie, Fei Huang, Huajun Chen, Ningyu Zhang
| Challenge: | Recent advances in Large Language Models (LLMs) have demonstrated remarkable progress in machine writing such as open domain long-form generation. |
| Approach: | They propose a slow-thinking machine writing framework that emulates the human-like process of iterative expansion and reflection to improve the knowledge density of generated articles. |
| Outcome: | The proposed framework improves the knowledge density of generated articles without compromising metrics such as coherence and depth. |
Wasserstein-Fisher-Rao Embedding: Logical Query Embeddings with Local Comparison and Global Transport (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for answering complex queries on knowledge graphs lack a local and global scoring function. |
| Approach: | They propose a convolution-based algorithm for linear time computation and a block diagonal kernel to enforce the trade-off between local and global embeddings. |
| Outcome: | The proposed model outperforms existing methods on standard datasets, evaluation sets with combinatorially complex queries, and hierarchical knowledge graphs. |
Enhancing LLM Capabilities Beyond Scaling Up (2024.emnlp-tutorials)
Copied to clipboard
| Challenge: | general-purpose large language models (LLMs) are expanding in scale and access to unpublic training data. |
| Approach: | This tutorial aims to examine the capabilities of general-purpose large language models . authors discuss adaptation of LLMs to address conflicts, defense against attacks . |
| Outcome: | This tutorial aims to examine the evolution of general-purpose large language models (LLMs) the authors argue that the evolution is dependent on the availability of training data and the scale of the models. |
Teaching Small Language Models Reasoning through Counterfactual Distillation (2024.emnlp-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) have demonstrated remarkable performance in a wide range of downstream tasks. |
| Approach: | They propose a counterfactual distillation framework that leverages LLMs to generate high-quality counterfacts and utilizes multi-view CoT to enhance the diversity of reasoning samples. |
| Outcome: | The proposed framework enhances reasoning capabilities of large language models and is more robust to OOD data. |
PaCE: Unified Multi-modal Dialogue Pre-training with Progressive and Compositional Experts (2023.acl-long)
Copied to clipboard
| Challenge: | Existing research on multi-modal dialogue pre-training is limited due to limited availability of multi-dimensional data . a recent emergence of chatGPT 1 has increased confidence in the potential for this goal . |
| Approach: | They propose a framework for multi-modal dialogue pre-training that integrates experts to accommodate multi-faceted tasks. |
| Outcome: | The proposed framework achieves state-of-the-art on eight multi-modal dialog benchmarks. |
CMMaTH: A Chinese Multi-modal Math Skill Evaluation Benchmark for Foundation Models (2025.coling-main)
Copied to clipboard
Zhongzhi Li, Ming-Liang Zhang, Pei-Jie Wang, Jian Xu, Rui-Song Zhang, Yin Fei, Zhi-Long Ji, Jin-Feng Bai, Zhen-Ru Pan, Jiaxin Zhang, Cheng-Lin Liu
| Challenge: | Large language models excel in various language tasks, while large multimodal models effectively handle visual-language problems. |
| Approach: | They propose to use a multimodal multimodal model evaluation benchmark to evaluate model performance in Chinese K12 classrooms. |
| Outcome: | The proposed model evaluation tool is integrated with the CMMaTH dataset. |
EvolveSearch: An Iterative Self-Evolving Search Agent (2025.emnlp-main)
Copied to clipboard
Ding-Chu Zhang, Yida Zhao, Jialong Wu, Liwen Zhang, Baixuan Li, Wenbiao Yin, Yong Jiang, Yu-Feng Li, Kewei Tu, Pengjun Xie, Fei Huang
| Challenge: | Existing approaches to enabling LLM web search proficiency struggle with data production in open-search domains, while supervised fine-tuning struggles with data utilization efficiency. |
| Approach: | They propose an iterative self-evolution framework that combines SFT and RL to enhance agentic web search capabilities without external human-annotated reasoning data. |
| Outcome: | EvolveSearch achieves 4.7% improvement over current state-of-the-art in seven benchmarks . supervised fine-tuning struggles with data production in open-search domains compared with RL . |
Distinguish Before Answer: Generating Contrastive Explanation as Knowledge for Commonsense Question Answering (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing knowledge-enhanced methods have trouble obtaining knowledge from different knowledge bases . a concept-centric model can be used to generate a contrastive explanation for QA tasks . |
| Approach: | They propose a Concept-centric Prompt-bAsed Contrastive Explanation Generation model which converts obtained symbolic knowledge into the contrastive explanation for better distinguishing the differences among given candidates. |
| Outcome: | The proposed model achieves new SOTA on CSQA, QASC, and OBQA. |
Geoparsing: Diagram Parsing for Plane and Solid Geometry with a Unified Formal Language (2026.findings-acl)
Copied to clipboard
Peijie Wang, Ming-Liang Zhang, Jun Cao, Chao Deng, Dekang Ran, Pi Bu, Hongda Sun, Xuan Zhang, Yingyao Wang, Jun Song, Bo Zheng, Fei Yin, Cheng-Lin Liu
| Challenge: | Recent advances in Multimodal Large Language Models (MLLMs) have demonstrated remarkable capabilities across various vision reasoning tasks. |
| Approach: | They propose a unified formal language that integrates plane and solid geometry, comprehensively covering geometric structures and semantic relations. |
| Outcome: | The proposed language achieves state-of-the-art parsing performance and significantly boosts MLLMs’ capabilities for downstream geometry reasoning tasks. |
WebWalker: Benchmarking LLMs in Web Traversal (2025.acl-long)
Copied to clipboard
Jialong Wu, Wenbiao Yin, Yong Jiang, Zhenglin Wang, Zekun Xi, Runnan Fang, Linhai Zhang, Yulan He, Deyu Zhou, Pengjun Xie, Fei Huang
| Challenge: | Large Language Models (LLMs) have demonstrated impressive capabilities across a wide range of natural language processing tasks. |
| Approach: | They propose a benchmark to assess the ability of LLMs to perform web traversal by using an explore-critic paradigm. |
| Outcome: | The proposed framework mimics human-like web navigation through an explore-critic paradigm and demonstrates the effectiveness of RAG combined with WebWalker in real-world scenarios. |
LANS: A Layout-Aware Neural Solver for Plane Geometry Problem (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing neural solvers take GPS as vision-language task but lack layout awareness . Existing models are criticized for complex rules and poor adaptability . |
| Approach: | They propose a layout-aware neural solver called LANS that integrates two modules to solve GPS. |
| Outcome: | The proposed solver outperforms existing neural and symbolic solvers on two datasets. |
LawBench: Benchmarking Legal Knowledge of Large Language Models (2024.emnlp-main)
Copied to clipboard
Zhiwei Fei, Xiaoyu Shen, Dawei Zhu, Fengzhe Zhou, Zhuo Han, Alan Huang, Songyang Zhang, Kai Chen, Zhixin Yin, Zongwen Shen, Jidong Ge, Vincent Ng
| Challenge: | LegalBench evaluated 20 LLMs in 162 legal tasks in 20 countries and jurisdictions. |
| Approach: | They present a comprehensive evaluation of 21 popular Large Language Models and the first comparative analysis of the empirical results. |
| Outcome: | The proposed benchmarks are based on the Bloom’s cognitive taxonomy and are compared to 21 popular LLMs. |
Optimizing Instruction Synthesis: Effective Exploration of Evolutionary Space with Tree Search (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Extensive research has highlighted the quality of instruction data is essential for the success of this alignment. |
| Approach: | They propose a framework for iteratively improving existing instruction data by using Monte Carlo tree search to find suitable prompts that align the language model to effectively learn multiple skills. |
| Outcome: | The proposed framework improves the evaluation scores of seed instruction data, raising the average evaluation scores from 2.19 to 3.81. |
The Butterfly Effect of Model Editing: Few Edits Can Trigger Large Language Models Collapse (2024.findings-acl)
Copied to clipboard
| Challenge: | Even a single edit can trigger model collapse, manifesting as significant performance degradation in various benchmark tasks. |
| Approach: | They propose to use perplexity as a surrogate metric to determine whether an edited model's performance is affected by a single edit. |
| Outcome: | The proposed method shows that even a single edit can cause model collapse, manifesting as significant performance degradation in various benchmark tasks. |
Contrastive Instruction Tuning (2024.findings-acl)
Copied to clipboard
| Challenge: | Current LLMs exhibit limited robustness to unseen instructions, generating inconsistent outputs when the same instruction is phrased with slightly varied forms or language styles. |
| Approach: | They propose a method which maximizes the similarity between the hidden representations of semantically equivalent instruction-instance pairs while minimizing the similarities between semantically different ones. |
| Outcome: | Experiments on the PromptBench benchmark show that Contrastive Instruction Tuning improves LLMs’ robustness to unseen instructions with variations across character, word, sentence, and semantic levels by +2.5% in accuracy. |
Extending Complex Logical Queries on Uncertain Knowledge Graphs (2025.acl-long)
Copied to clipboard
| Challenge: | Existing studies on logical queries on knowledge graphs overlook the incompleteness of KGs. |
| Approach: | They propose an ML-based approach to answer soft queries on uncertain knowledge . they propose to use forward inference and backward calibration to avoid catastrophic errors . |
| Outcome: | The proposed method ensures there are no catastrophic cascading errors while maintaining the same complexity as state-of-the-art inference algorithms for first-order queries. |
GeoEval: Benchmark for Evaluating LLMs and Multi-Modal Models on Geometry Problem-Solving (2024.findings-acl)
Copied to clipboard
| Challenge: | Recent advances in large language models (LLMs) and multi-modal models (MMs) have demonstrated remarkable capabilities in problem-solving, but their proficiency in tackling geometry math problems has not been thoroughly evaluated. |
| Approach: | They propose a benchmark to evaluate the performance of large language models and multi-modal models in solving geometry math problems. |
| Outcome: | The proposed model achieves 55.67% accuracy on main subset but only 6.00% accuracy on hard subset. |
"I Don’t Know What to Say": A Fact-Filling Questionnaire Method to Help Non-Experts Talk to LegalAI Assistant (2026.findings-acl)
Copied to clipboard
| Challenge: | Recent advances in large language models (LLMs) have greatly expanded the scope of legal AI. |
| Approach: | They propose a method that generates questionnaires to help users refine queries . they leverage an iterative training process that collects valuable questionnaires . |
| Outcome: | The proposed method improves the completeness of queries and ensures the performance of domain-specific models in downstream legal tasks. |
BrowseConf: Confidence-Guided Test-Time Scaling for Web Agents (2026.findings-acl)
Copied to clipboard
Litu Ou, Kuan Li, Huifeng Yin, Liwen Zhang, Zhongwang Zhang, Xixi Wu, Rui Ye, Zile Qiao, Yong Jiang, Pengjun Xie, Fei Huang, Jingren Zhou
| Challenge: | Existing work on confidence in LLMs is limited. |
| Approach: | They propose to use confidence scores to determine model answer quality and encourage model to try again until it reaches satisfactory confidence level. |
| Outcome: | The proposed methods significantly reduce token consumption while demonstrating competitive performance compared to baseline fixed budget methods. |
OS Agents: A Survey on MLLM-based Agents for Computer, Phone and Browser Use (2025.acl-long)
Copied to clipboard
Xueyu Hu, Tao Xiong, Biao Yi, Zishu Wei, Ruixuan Xiao, Yurun Chen, Jiasheng Ye, Meiling Tao, Xiangxin Zhou, Ziyu Zhao, Yuhuai Li, Shengze Xu, Shenzhi Wang, Xinchen Xu, Shuofei Qiao, Zhaokai Wang, Kun Kuang, Tieyong Zeng, Liang Wang, Jiwei Li, Yuchen Eleanor Jiang, Wangchunshu Zhou, Guoyin Wang, Keting Yin, Zhou Zhao, Hongxia Yang, Fan Wu, Shengyu Zhang, Fei Wu
| Challenge: | a new generation of (M)LLMs is enabling the creation of superintelligent AI assistants . OS Agents can complete tasks autonomously and have the potential to significantly enhance the lives of billions of users worldwide. |
| Approach: | They propose to build OS Agents that operate within operating systems' GUIs and GUIs . they examine evaluation metrics and benchmarks to identify promising directions . |
| Outcome: | The proposed agents are based on operating systems (OS) and operating systems frameworks. |
Meta Distant Transfer Learning for Pre-trained Language Models (2021.emnlp-main)
Copied to clipboard
| Challenge: | Notable PLMs are available for text classification tasks, but performance of PLM on downstream tasks may be limited by the availability of training set. |
| Approach: | They propose a meta-learning framework to learn the transferable knowledge across tasks using PLMs. |
| Outcome: | The proposed framework outperforms baselines on seven datasets and is task-agnostic and unbiased. |
CBLUE: A Chinese Biomedical Language Understanding Evaluation Benchmark (2022.acl-long)
Copied to clipboard
Ningyu Zhang, Mosha Chen, Zhen Bi, Xiaozhuan Liang, Lei Li, Xin Shang, Kangping Yin, Chuanqi Tan, Jian Xu, Fei Huang, Luo Si, Yuan Ni, Guotong Xie, Zhifang Sui, Baobao Chang, Hui Zong, Zheng Yuan, Linfeng Li, Jun Yan, Hongying Zan, Kunli Zhang, Buzhou Tang, Qingcai Chen
| Challenge: | a new benchmark for biomedical language understanding is being developed in Chinese . most benchmarks are limited to English, which makes it difficult to replicate success in other languages. |
| Approach: | They propose to use Chinese biomedical language understanding evaluation benchmarks to evaluate Chinese models. |
| Outcome: | The proposed benchmarks show that the current models perform worse than the human ceiling. |
The Fall of ROME: Understanding the Collapse of LLMs in Model Editing (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Recent studies have found that model editing methods can cause large language models to collapse with just a single edit. |
| Approach: | They propose a method that uses prefixed keys and adds prefixes during testing to prevent model collapse. |
| Outcome: | The proposed method prevents model collapse while maintaining effectiveness, the authors show . Rank-One Model Editing (ROME) has been found to cause model collapse with just a single edit . |
DiningBench: A Hierarchical Multi-view Benchmark for Perception and Reasoning in the Dietary Domain (2026.acl-long)
Copied to clipboard
| Challenge: | Existing vision-language models lack fine-grained classification, single-view imagery, and inaccurate metadata. |
| Approach: | They propose a hierarchical, multi-view benchmark to evaluate VLMs across three levels of cognitive complexity. |
| Outcome: | The proposed benchmark evaluates vision-language models across three levels of complexity . it systematically identifies five primary failure modes . the proposed benchmarks are available on https://github.com/meituan/DiningBench. |