CV
Professional Summary
My research focuses on agentic coding and large language models, including reliable code reproduction from academic papers, tool-augmented reasoning, memory management, and generating synthetic trajectories for supervised fine-tuning and reinforcement learning on complex coding tasks.
Education
Ph.D. in Computer Science.
Supervised by Prof. Yulan He with Dr. Lin Gui as co-advisor. Research topics: large language models, in-context learning, interpretability.
M.S. in Software Engineering.
Average score: 86.42/100. Supervised by Prof. Deyu Zhou with research on question answering and code generation.
B.E. in Computer Science and Technology.
Average score: 90.10/100. Coursework highlights: linear algebra, advanced mathematics, probability theory, data structures, Java programming, software engineering, internet protocols.
Awards and Honors
- NMES International Studentship (2023 – 2027)
- Outstanding Graduate Student, Hefei University of Technology (2020)
- Merit Student, Hefei University of Technology (2018)
Work Experience
- Meta — Contractor (Nov. 2025 – Dec. 2025)
Worked with Dr. Yoram Bachrach on AI Scientist agents, LLM-based systems that automate research workflows and machine learning engineering tasks. - AstraZeneca — Internship (Jul. 2025 – Oct. 2025)
Supervised by Dr. Saseendran and Dr. Jin. Investigated generating multiple tokens per decoding step in Diffusion Language Models to accelerate decoding.
Invited Talks
- Meta, LLaMA Community Meet-up (Apr. 6, 2025): “Towards Automatic Code Reproduction for Scientific Papers: Benchmarks and Methodologies.”
Competitions
- National First Prize (Top 0.65%), China Undergraduate Mathematical Contest in Modelling (2018).
- 1st Place, Spider Leaderboard (2022). Model G3R ranked first on the “exact set match without values” metric and is currently fifth overall.
Publications
- SciReplicate-Bench: Benchmarking LLMs in Agent-driven Algorithmic Reproduction from Research Papers [arXiv]
Yanzheng Xiang, Hanqi Yan, Shuyin Ouyang, Lin Gui, Yulan He
In COLM 2025. - G3R: A Graph-Guided Generate-and-Rerank Framework for Cross-domain Text-to-SQL Generation [paper]
Yanzheng Xiang, Qian-Wen Zhang, Xu Zhang, Zejie Liu, Yunbo Cao, Deyu Zhou
In Findings of ACL 2023. - Addressing Order Sensitivity of In-Context Demonstration Examples in Causal Language Models [arXiv]
Yanzheng Xiang, Hanqi Yan, Lin Gui, Yulan He
In Findings of ACL 2024. - The Mystery of In-Context Learning: A Comprehensive Survey on Interpretation and Analysis [paper]
Yuxiang Zhou, Jiazheng Li, Yanzheng Xiang, Hanqi Yan, Lin Gui, Yulan He
In EMNLP 2024. - A Divide-And-Conquer Approach for Multi-label Multi-hop Relation Detection in Knowledge Base QA [paper]
Deyu Zhou, Yanzheng Xiang, Linhai Zhang, Chenchen Ye, Qian-Wen Zhang, Yunbo Cao
In Findings of EMNLP 2021. - PECAN: LLM-Guided Dynamic Progress Control with Attention-Guided Hierarchical Weighted Graph for Long-Document QA [paper]
Xinyu Wang, Yanzheng Xiang, Lin Gui, Yulan He
In Findings of ACL 2025. - Encourage or Inhibit Monosemanticity? Revisit Monosemanticity from a Feature Decorrelation Perspective [paper]
Hanqi Yan, Yanzheng Xiang, Guangyi Chen, Yifei Wang, Lin Gui, Yulan He
In EMNLP 2024.
Talks
Invited Talk, Meta, LLaMA Community Meet-up, London, United Kingdom, April 06, 2025
I presented our latest work on SciReplicate-Bench and shared methodologies for building agentic LLM systems that can reliably reproduce code from scientific publications. The talk covered benchmarking strategies, memory management, and tooling considerations for research automation.
Conference proceedings talk, Testing Institute of America 2014 Annual Conference, Los Angeles, CA, March 01, 2014
This is a description of your conference proceedings talk, note the different field in type. You can put anything in this field.
Tutorial, UC-Berkeley Institute for Testing Science, Berkeley CA, USA, March 01, 2013
More information here
Talk, UC San Francisco, Department of Testing, San Francisco, California, March 01, 2012
This is a description of your talk, which is a markdown files that can be all markdown-ified like any other post. Yay markdown!
Teaching
Workshop, University 1, Department, 2015
This is a description of a teaching experience. You can use markdown like any other post.
Undergraduate course, University 1, Department, 2014
This is a description of a teaching experience. You can use markdown like any other post.
Other Skills
- IT: C++, SQL, Python, LaTeX
- Languages: English (Fluent, IELTS 7.0), Chinese (Native)
