Yuexiang Xie
Tongyi Lab, Alibaba Group
Email: yuexiang.xyx AT alibaba-inc.com
I am a research scientist at Tongyi Lab, Alibaba Group, and a member of the AgentScope-ai team, where we build and maintain open-source projects such as AgentScope and QwenPaw, among others. I received my Master's Degree from Peking University in 2020, advised by Prof. Ying Shen and Prof. Kai Lei.
My research focuses on large language models (LLMs) and agent systems, with the goal of bridging academic advances and real-world applications. As first author or industrial mentor, I have published 40+ papers (2,000+ citations) at top venues such as ICLR, ICML, NeurIPS, VLDB, SIGMOD, KDD, ACL, EMNLP, AAAI, SIGIR, WWW, etc., covering topics spanning LLM-based agents, multi-agent systems, reinforcement learning for LLMs, LLM alignment, federated learning, knowledge graphs, and AutoML.
We are building the next generation of real-world agent applications — from research ideas to open-source projects used by thousands. If you are excited about LLM agents, open-source, and publishing at top venues, feel free to drop me an email. Internship and collaboration positions are open.
Publications
-
A Practical Introduction to Federated Learning
Awards & Scholarships
- Best Paper Best Paper Award at Applied Data Science Track, KDD 2022
- Competition KDD Cup, AutoML Track, 4/149, May 2020
- Scholarship National Scholarship, Sep 2014
Featured Projects
More projects at github.com/agentscope-ai
Academic Service
Regular Reviewer / PC Member: ICLR, ICML, NeurIPS, NeurIPS Datasets and Benchmarks, ACL ARR, KDD, WWW, AAAI, TNNLS, TBD