About Me
I am a Ph.D. student in the Knowledge Engineering Group (KEG), Department of Computer Science and Technology, Tsinghua University, advised by Prof. Juanzi Li. Before starting my Ph.D., I received my M.S. in Computer Science and Technology from Tsinghua University in 2025, and my dual bachelor’s degrees in Information Management and Information Systems and Software Engineering from Tsinghua University in 2023.
My research focuses on large language models, with a recent focus on LLM agents for scientific research.
Currently, I have 3 first-author papers and 8 co-authored papers, with 350+ Google Scholar citations. You can find my publications on Google Scholar.
Research Interests
My vision is to build powerful and responsible AI agents that augment human intelligence, productivity, and well-being. My recent work centers on two directions:
1) LLM Agents for Scientific and Strategic Research
Developing capable agentic systems that accelerate scientific research and strategic decision making. Representative works: EurekAgent, StockBench.
2) General LLM Agents and Model Understanding
Building and evaluating general-purpose agentic systems across complex tasks and environments, while studying the capability boundaries and internal mechanisms of LLMs. Representative works: AtomR, LLMAEL, AGENTIF, KoLA.
News
- 🎉 [Jun 2026] We open-sourced EurekAgent, won the BAAI Agent for Science Competition Highlight Award, and gave an invited oral presentation at the 2026 BAAI Conference.
- 🔥 [May 2026] Our paper AISE-Bench: A Full-Cycle Curated Benchmark for Information Seeking on Academic Knowledge Graphs was accepted to KDD 2026.
- 🌟 [Dec 2025] Our papers AGENTIF: Benchmarking Instruction Following of Large Language Models in Agentic Scenarios and How do Transformers Learn Implicit Reasoning? were both accepted to NeurIPS 2025 as apotlight papers.
- 📣 [Nov 2025] Our paper LLMAEL: Large Language Models are Good Context Augmenters for Entity Linking was accepted to CIKM 2025 as an oral paper.
- 📈 [Oct 2025] We open-sourced StockBench, a contamination-free benchmark for evaluating LLM agents in realistic, multi-month stock trading environments.
- 🚀 [Aug 2025] Our paper AtomR: Atomic Operator-Empowered Large Language Models for Heterogeneous Knowledge Reasoning was accepted to KDD 2025 as an oral paper.
Selected Publications
* Equal contribution. Please refer to my Google Scholar profile for the full list.
LLM Agents for Scientific and Strategic Research
EurekAgent: Agent Environment Engineering is All You Need For Autonomous Scientific Discovery
Amy Xin, Jiening Siow, Junjie Wang, Zijun Yao, Fanjin Zhang, Jian Song, Lei Hou, Juanzi Li
arXiv preprint, 2026. Award Winner, BAAI Agent for Science Competition; Invited Oral Presentation at 2026 BAAI ConferenceStockBench: Can LLM Agents Trade Stocks Profitably In Real-world Markets?
Yanxu Chen*, Zijun Yao*, Yantao Liu*, Amy Xin, Jin Ye, Jianing Yu, Lei Hou, Juanzi Li
arXiv preprint, 2025.AISE-Bench: A Full-Cycle Curated Benchmark for Information Seeking on Academic Knowledge Graphs
Fanjin Zhang*, Zhengyang Wang*, Ruixuan Huang, Kefan Zhang, Amy Xin, Yuanchun Wang, Shu Zhao, Evgeny Kharlamov, Jie Tang, Juanzi Li
KDD, 2026.
General LLM Agents and Model Understanding
AtomR: Atomic Operator-Empowered Large Language Models for Heterogeneous Knowledge Reasoning
Amy Xin*, Jinxin Liu*, Zijun Yao, Zhicheng Lee, Shulin Cao, Lei Hou, Juanzi Li
KDD, 2025. Oral PresentationLLMAEL: Large Language Models are Good Context Augmenters for Entity Linking
Amy Xin*, Yunjia Qi*, Zijun Yao, Fangwei Zhu, Kaisheng Zeng, Bin Xu, Lei Hou, Juanzi Li
CIKM, 2025. Oral PresentationAGENTIF: Benchmarking Instruction Following of Large Language Models in Agentic Scenarios
Yunjia Qi*, Hao Peng*, Xiaozhi Wang, Amy Xin, Youfeng Liu, Bin Xu, Lei Hou, Juanzi Li
NeurIPS, 2025. SpotlightVisKoP: Visual Knowledge Oriented Programming for Interactive Knowledge Base Question Answering
Zijun Yao*, Yuanyong Chen*, Xin Lv, Shulin Cao, Amy Xin, Jifan Yu, Hailong Jin, Jianjun Xu, Peng Zhang, Lei Hou, Juanzi Li
ACL, 2023. Best Demo AwardHow do Transformers Learn Implicit Reasoning?
Jiaran Ye*, Zijun Yao*, Zhidian Huang, Liangming Pan, Jinxin Liu, Yushi Bai, Amy Xin, Weichuan Liu, Xiaoyin Che, Lei Hou, Juanzi Li
NeurIPS, 2025. SpotlightKoLA: Carefully Benchmarking World Knowledge of Large Language Models
Jifan Yu*, Xiaozhi Wang*, Shangqing Tu, Shulin Cao, Daniel Zhang-Li, Xin Lv, Hao Peng, Zijun Yao, Xiaohan Zhang, Hanming Li, Chunyang Li, Zheyuan Zhang, Yushi Bai, Yantao Liu, Amy Xin, …, Lei Hou, Zhiyuan Liu, Bin Xu, Jie Tang, Juanzi Li
ICLR, 2024.
Open-source Projects
EurekAgent
An autonomous research system for metric-driven scientific discovery. Given a problem and evaluation criteria, EurekAgent coordinates CLI agents (e.g., Claude Code) to propose approaches, implement code, run experiments, and iterate toward stronger results, with Docker-based isolation, resumable runs, and easy human oversight.Awesome AI for Research
A curated collection of recent and significant AI-for-research papers, systems, benchmarks, and projects. The repository organizes work across research stages, agent roles, application domains, and evaluation benchmarks to help researchers navigate the evolving AI4Research landscape.
Honors and Awards
- BAAI Agent for Science Competition Highlight Award for EurekAgent (2026).
- ACL Best Demo Award for VisKoP (2023).
- Chinese Government Scholarship, awarded for 4 consecutive years (2023-2026).
- Tsinghua University Merit Scholarships, awarded three times (2021-2023).
Activities and Hobbies
🎵 I have a deep passion for music, spanning songwriting, vocals, and instruments. I am a music content creator with 100K+ followers, 100M+ video views, and 80M+ audio streams across platforms. I won the 31st Tsinghua University Campus Singing Competition with an original song.
You can find my music channels here: 📺 Bilibili and ▶️ YouTube.
🏃♀️ Feel free to follow me on 📕 Xiaohongshu and 𝕏 for recent updates.
📚 I also enjoy reading philosophy, sociology, psychology, and economics.
Last updated: July 2026.