About Me

I am a Ph.D. candidate in Computer Science at Tianjin University, China, since Fall 2024, supervised by Prof. Qinghua Hu. I received an MEng in Machine Learning and Computer Vision from the Australian National University in 2020 and a BSc in Hydrology and Water Resources Engineering from Wuhan University in 2017.

My research interests lie in online learning, test time reinforcement learning, and LLM-Based agents, with a focus on both theoretical foundations and practical deployment of adaptive decision-making systems.

News

• May 2026 Three papers accepted to ICML 2026: ALSO (first author), T-POP, and Social Hippocampus Memory Learning.
• Sep 2025 Joined CUHKSZ as a visiting student under the supervision of Prof. Zhongxiang Dai, engaging in research on prompt optimization for multi-agent systems.

Publications

Bold highlights Xiang Li; † denotes corresponding author.

1. ALSO: Adversarial Online Strategy Optimization for Social Agents.

   Xiang Li, Liping Yi, Mingze Kong, Min Zhang, Zhongxiang Dai, Qinghua Hu

   ICML 2026[PDF][Code][Project]

2. Social Hippocampus Memory Learning.

   Liping Yi, Zhiming Zhao, Kewen Zhu, Xiang Li, Zhiwei Shang, Qinghua Hu

   ICML 2026

3. T-POP: Test-Time Personalization with Online Preference Feedback.

   Zikun Qu, Min Zhang, Mingze Kong, Xiang Li, Zhiwei Shang, Zhiyong Wang, Yikun Ban, Shuang Qiu, Yao Shu, Zhongxiang Dai

   ICML 2026 [PDF]

Papers & Preprints

Bold highlights Xiang Li; † denotes corresponding author.

1. Workflow-R1: Group Sub-sequence Policy Optimization for Multi-turn Workflow Construction.

   Mingze Kong, Zikun Qu, Zhongquan Zhou, Pengyu Liang, Xiang Li, Zhiwei Shang, Zhi Hong, Kaiyu Huang, Zhiyong Wang, Zhongxiang Dai

   arXiv preprint arXiv:2602.01202, 2026 [PDF]

2. Large Model Driven Solar Activity AI Forecaster: A Scalable Dual Data-Model Framework.

   Jingjing Wang, Pengyu Liang, Tingyu Wang, Ming Li, Yanmei Cui, Siwei Liu, Xin Huang, Xiang Li, Minghui Zhang, Yunshi Zeng, Zhu Cao, Jiekang Feng, Qinghua Hu, Bingxian Luo, Bing Cao

   arXiv preprint arXiv:2508.06892, 2025 [PDF]

Projects

Research isn't everything—I always enjoy building things that are simply fun.

• 

  AIPPT (Automated PPT Generator)

  2025

  A fully on-premises system that transforms structured briefing notes into polished slide decks with customizable branding, layout templates, and collaborative review workflows.

• 

  Liandanlu (Large-Scale Compute Scheduler)

  2024 – Present

  Co-developed with Jiekang Feng. Liandanlu provides a plug-and-play GPU scheduling experience with multi-node awareness, fine-grained resource quotas, and rapid task orchestration for research labs.

• Cyber-Autonomous Swarm Simulation Platform

  2023 – 2024

  Co-developed with Xiaowei Zhao. The platform offers high-fidelity multi-agent simulation, swarm behavior scripting, and fast iteration for testing autonomous strategies.

• Automated Depalletizing

  2023

  Implemented mixed-pallet detection and grasp sequencing to streamline warehouse depalletizing with minimal manual supervision.

• Bin Picking System

  2022

  Developed perception and grasp planning pipelines that let articulated robot arms reliably pick cluttered industrial parts from bins.

• Primitive Fitting Toolkit

  2021

  Built a 3D primitive fitting pipeline that segments scene geometry into cylinders, planes, and cuboids for downstream CAD reconstruction.