Welcome to Yifan Wei (Eric)’s Personal Homepage!
I am currently pursuing my PhD in the School of Computer Science and Engineering at Beihang University, under the supervision of Professor Angsheng Li. I had a wonderful time at the Institute of Automation, Chinese Academy of Sciences as a master’s student, where I was advised by Professor Kang Liu and co-advised by Professor Jun Zhao.
My research interests lie in the field of natural language processing (NLP), with a focus on Large Language Models (LLMs), including Question Answering and Knowledge Editing. Currently, I am highly intrigued by exploring the intersection of LLMs and AI safety.
To date, I have authored three papers as the first author, which have been published at AI conferences including the CIKM, EMNLP, IJCAI, and AAAI,
with total google scholar citations ().
“Perception brings meaning to sensation, so perception produces and interpretation of the external world, not a perfect representation of it.”
「知觉赋予感觉以意义,因此知觉产生的是对世界的解释,而不是对世界的完美表征。」
🔥 News
- 2025.4: 📢📢📢 Our new paper “Structural Entropy Guided Agent for Detecting and Repairing Knowledge Deficiencies in LLMs” is now available on arXiv!
- 2025.4: 🎉 Our paper “SetKE: Knowledge Editing for Knowledge Elements Overlap” has been accepted to IJCAI 2025!
- 2025.2: 🎉 Our paper “Towards Effective, Efficient and Unsupervised Social Event Detection in the Hyperbolic Space” has been accepted to AAAI 2025!
- 2024.09: 🎉🎉 Our paper “Neeko: Leveraging Dynamic LoRA for Efficient Multi-Character Role-Playing Agent” has been accepted to EMNLP 2024!
- 2024.07: 🎉🎉 Our paper “DAMe: Personalized Federated Social Event Detection with Dual Aggregation Mechanism” has been accepted to CIKM 2024!
- 2024.07: 🎉🎉 Our paper “Does Knowledge Localization Hold True? Surprising Differences Between Entity and Relation Perspectives in Language Models” has been accepted to CIKM 2024 (short paper)!
- 2024.05: 📢 Our paper “EX-FEVER: A Dataset for Multi-hop Explainable Fact Verification” has been accepted to ACL 2024 Findings!
- 2023.10: 🎉 Our paper “MenatQA: A New Dataset for Testing the Temporal Comprehension and Reasoning Abilities of Large Language Models” has been accepted to EMNLP 2023 Findings!
📝 Publications

Structural Entropy Guided Agent for Detecting and Repairing Knowledge Deficiencies in LLMs
Yifan Wei, Xiaoyan Yu, Tengfei Pang, Angsheng Li, Li Du
we propose a novel Structural Entropy-guided Knowledge Navigator (SENATOR) framework that addresses the intrinsic knowledge deficiencies of LLMs.

SetKE: Knowledge Editing for Knowledge Elements Overlap
Yifan Wei, Xiaoyan Yu, Ran Song, Angsheng Li
we propose a new formulation, Knowledge Set Editing (KSE), and introduce SetKE, a method that edits sets of triplets simultaneously.

Towards Effective, Efficient and Unsupervised Social Event Detection in the Hyperbolic Space
Xiaoyan Yu, Yifan Wei, Shuaishuai Zhou, Liehuang Zhu, Philip S. Yu
We introduce HyperSED, an effective and efficient unsupervised framework for social event detection, which exploits the benefits of graph learning in the hyperbolic space and structural information to achieve event detection.

Neeko: Leveraging Dynamic LoRA for Efficient Multi-Character Role-Playing Agent
Xiaoyan Yu, Tongxu Luo, Yifan Wei *, Fangyu Lei, Yiming Huang, Hao Peng, Liehuang Zhu
We present Neeko, an innovative framework designed for efficient multiple characters imitation. Unlike existing methods, Neeko employs a dynamic low-rank adapter (LoRA) strategy, enabling it to adapt seamlessly to diverse characters.

DAMe: Personalized Federated Social Event Detection with Dual Aggregation Mechanism
Xiaoyan Yu, Yifan Wei, Pu Li, Shuaishuai Zhou, Hao Peng, Li Sun, Liehuang Zhu, Philip S Yu
This paper proposes a personalized federated learning framework with a dual aggregation mechanism for social event detection, namely DAMe.

Yifan Wei, Xiaoyan Yu, Yixuan Weng, Huanhuan Ma, Yuanzhe Zhang, Jun Zhao, Kang Liu
This study investigates the differences between entity and relational knowledge through knowledge editing. Our findings reveal that entity and relational knowledge cannot be directly transferred or mapped to each other.

EX-FEVER: A Dataset for Multi-hop Explainable Fact Verification
Huanhuan Ma, Weizhi Xu, Yifan Wei, Liuji Chen, Liang Wang, Qiang Liu, Shu Wu, Liang Wang
We introduce a large scale Multi-hop fact checking dataset with textual explanations, which can be used to evaluate the explainability of fact verification models.

Yifan Wei, Yisong Su, Huanhuan Ma, Xiaoyan Yu, Fangyu Lei, Yuanzhe Zhang, Jun Zhao, Kang Liu
Findings of the Association for Computational Linguistics: EMNLP 2023
We construct Multiple Sensitive Factors Time QA (MenatQA), which encompasses three temporal factors (scope factor, order factor, counterfactual factor) with total 2,853 samples for evaluating the time comprehension and reasoning abilities of LLMs.

Assessing knowledge editing in language models via relation perspective
Yifan Wei, Xiaoyan Yu, Huanhuan Ma, Fangyu Lei, Yixuan Weng, Ran Song, Kang Liu

S3HQA: A Three-Stage Approach for Multi-hop Text-Table Hybrid Question Answering
Fangyu Lei, Xiang Li, Yifan Wei, Shizhu He, Yiming Huang, Jun Zhao, Kang Liu
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics

Multi-View Graph Representation Learning for Answering Hybrid Numerical Reasoning Questions
Yifan Wei, Fangyu Lei, Yuanzhe Zhang, Jun Zhao, Kang Liu
🚀 Projects
- Neeko: Leveraging Dynamic LoRA for Efficient Multi-Character Role-Playing Agent.
- TableQAKit: A Toolkit for Table-based Question Answering.
📖 Educations
-
2021.09 - 2024.07, M.S. in Artificial Intelligence, Institute of Automation, Chinese Academy of Sciences. Advisors: Prof. Kang Liu and Prof. Jun Zhao.
-
2024.09 - , Ph.D. candidate at the School of Computer Science and Engineering, Beihang University.
💻 Internships
📅 Academic Services
📖 Reviewers
- Annual Meeting of the Association for Computational Linguistics (ACL,EMNLP,NAACL), Reviewer
- Annual Conference on Neural Information Processing Systems (NeurIPS), Reviewer
- ACM International Conference on Information and Knowledge Management (CIKM), PC member
- International Joint Conference on Artificial Intelligence (IJCAI), PC member
- The Association for the Advancement of Artificial Intelligence (AAAI), PC member