Lingyun (Richard) Xiao

Lingyun (Richard) Xiao 肖凌云

I'm a PhD student in Operations Research at The University of Texas at Austin, where I'm honored to be advised by Prof. Peter Stone in the Learning Agents Research Group (LARG).

I study how AI agents coordinate with other agents and with humans when there is no central planner: coding agents serving different developers in one codebase, LLM agents teaming up with partners they've never met, and humanoid service robots helping people at home. My work draws on operations research, reinforcement learning, and large language models, and spans both software agents and physical robots.

Before UT, I received my M.S. in Industrial and Systems Engineering from UW–Madison and my B.A. in Applied Mathematics and Economics from UC Berkeley. I also spent a summer as a Revenue Management Pricing Operations Research Intern at American Airlines.

Email / CV / Scholar / X / LinkedIn / GitHub

News

Nov 2026 — ParallACT is out on arXiv, with a project page and dataset explorer.
Sep 2026 — JaxAHT preprint released, along with library v1.1.
Dec 2025 — Our paper on the Concordia contest appeared at NeurIPS 2025 (Datasets and Benchmarks Track).
Jul 2025 — UT Austin Villa @Home won 2nd place at the RoboCup@Home International Competition in Salvador, Brazil.

Research

Representative papers are below. Highlighted papers are ones I led. * denotes equal contribution.

	ParallACT: Decentralized Multi-Agent Coordination and Communication in Shared Codebases Rishabh Mediratta, Lingyun Xiao, Peter Stone, Joydeep Biswas, Elias Stengel-Eskin arXiv, 2026 project page / arXiv / code & data Two coding agents, two developers' requests, one codebase. Each agent's change looks fine on its own, but the combined code often fails, even when the agents can talk.
	JaxAHT: A JAX-Based Library for Ad Hoc Teamwork Caroline Wang, Rolando Fernandez, Zelal Su Mustafaoglu, Montek Singh Kundan, Johnny Liu, Aditya Madhan, Zhihan Wang, Lingyun Xiao, Di Yang Shi, Jiaxun Cui, Arrasy Rahman, Peter Stone ICLR Workshop on Multi-Agent Learning and Generative AI (MALGAI), 2026 project page / arXiv / code An open-source JAX library for ad hoc teamwork, where agents must coordinate with partners they have never seen before.
	Evaluating Generalization Capabilities of LLM-Based Agents in Mixed-Motive Scenarios Using Concordia Chandler Smith, Marwa Abdulhai, Manfred Diaz, …, Lingyun Xiao, …, Natasha Jaques, Tim Baarslag, Jose Hernandez-Orallo, Joel Z. Leibo NeurIPS Datasets and Benchmarks, 2025 arXiv Results from the NeurIPS 2024 Concordia Contest, testing whether LLM agents can find mutual gains with new partners across scenarios from negotiation to collective action.
	Online Nonparametric Process Monitoring for IoT Systems Using Edge Computing Ziqian Zheng, Jiahui Zhang, Lingyun Xiao, Warren R. Williams, Jing-Ru C. Cheng, Kaibo Liu IISE Transactions, 2024 paper Nonparametric monitoring of multi-sensor IoT data streams computed at the edge to detect anomalies online.