|
I am a third-year Master student at the Wangxuan Institute of Computer Technology, Peking University, supervised by Prof. Xiaojun Wan. Preivously, I received my bachelor's degree from Peking University in 2022. I worked as an intern at Microsoft. Currently, I am also a visiting student at the UCSB NLP Group, advised by Prof. William Yang Wang. Research: I aim to make models more generalizable across real-world tasks. I have focused on evaluating and analyzing the generalization capabilities of models, exploring their knowledge boundaries, and designing methods for knowledge editing, self-alignment, and self-referential agents to improve model versatility and performance. I believe in using methods that are simple, generalizable, and graceful. I am seeking 2025 Fall CS Ph.D. opportunities. Feel free to contact me if you are interested in my research or want to discuss with me. |
|
|
|
|
New! Gödel Agent: A Self-Referential Agent Framework for
Recursive Self-Improvement
|
|
ContraSolver: Self-Alignment of Language Models by Resolving Internal Preference Contradictions
|
|
|
|
New! Evaluating Self-Generated Documents for Enhancing
Retrieval-Augmented Generation with LLMs
|
|
New! ChemAgent: Self-updating Memories in LLMs Improves
Chemical Reasoning
Xiangru Tang, Tianyu Hu, Muyang Ye, Yanjun Shao, Xunjian Yin, Siru Ouyang, Wangchunshu Zhou,
Pan Lu, Zhuosheng Zhang, Yilun Zhao, Arman Cohan, Mark Gerstein
ICLR 2025 [ Paper ] |
|
DSGram: Dynamic Weighting Sub-Metrics for Grammatical
Error Correction in the Era of Large Language Models
|
|
Themis: A Reference-free NLG Evaluation Language Model
with Flexibility and Interpretability
|
|
Benchmarking Knowledge Boundary for Large Language Model: A Different Perspective on Model
Evaluation
|
|
History Matters: Temporal Knowledge Editing in Large Language Model
|
|
Error-Robust Retrieval for Chinese Spelling Check
|
|
ALCUNA: Large Language Models Meet New Knowledge
|
|
Exploring Context-Aware Evaluation Metrics for Machine Translation
|
|
How Do Seq2Seq Models Perform on End-to-End Data-to-Text Generation?
|
|
|
|
|
|
|
Thanks Jon Barron and Xueqing Wu for this amazing work. |