Jiayuan Su

I am an Applied Research Staff at WeChat, Tencent, where I work on agentic AI, large language models, and recommender systems. Currently, my interests include applying agentic LLMs to build self-evolving systems, such as recommender systems, and exploring tool-use and skill-use capabilities for intelligent agents across both recommendation and general domains.

Previously, I earned my Master's degree from Zhejiang University and had the opportunity to intern at Microsoft Research Asia, the Tsinghua NLP Lab and so on.

I'm always happy to connect with people interested in LLMs, AI agents, or machine learning in general. Feel free to reach out!

Keywords: Agentic AI · LLMs · Recommender Systems

News

[Apr 2026] One paper accepted by ACL 2026.
[Mar 2026] Graduated from Zhejiang University with a Master's degree.
[Dec 2025] Awarded the National Scholarship.
[Nov 2025] Two papers accepted by AAAI 2026 Oral. See you in Singapore.
[Aug 2025] Two papers accepted by EMNLP 2025.
[May 2025] One paper accepted by ACL 2025.
[Sep 2024] One paper accepted by EMNLP 2024.

Selected Papers

CP-Router: An Uncertainty-Aware Router Between LLM and LRM [link]

In AAAI 2026
SheetBrain: A Neuro-Symbolic Agent for Accurate Reasoning over Complex and Large Spreadsheets [link]

In AAAI 2026
MT-RewardTree: A Comprehensive Framework for Advancing LLM-Based Machine Translation via Reward Modeling [link]

In EMNLP 2025
M-MAD: Multidimensional Multi-Agent Debate for Advanced Machine Translation Evaluation [link]

In ACL 2025
API Is Enough: Conformal Prediction for Large Language Models Without Logit-Access [link]

In EMNLP 2024

Internships

2026 Xiaohongshu (Engineering)
2025 Tsinghua NLP Lab → Microsoft Research Asia (Applied Research)
2024 KAUST (Research)
2023 University of Illinois Chicago (Research)