I am a master's student in Computer Technology at South China University of Technology, advised by Prof. Tong Zhang. Before that, I received my bachelor's degree from Jinan University.

My research mainly focuses on multimodal understanding, vision grounding, agent systems, multi-turn agent systems, post-training, and DPO. I have published first-author research papers on multimodal understanding at ICML and ICLR.

I am currently exploring opportunities to pursue PhD programs beginning in Fall 2027 to further my academic journey. If you have any relevant openings or academic collaborations, please feel free to contact me at 1204385321@qq.com.

šŸ“ŒNews

  • šŸŽ‰ 2026.05.01: Steer Where It Matters was accepted by ICML 2026.
  • šŸŽ‰ 2026.01.27: P2-DPO was accepted as an ICLR 2026 Poster.
  • 2026.05: Started an internship at ByteDance Douyin Game, working on complex agent systems and agentic post-training.
  • 2025.09: Started an internship at China Southern Power Grid AI Company, working on GUI agents.
  • 2025.03: Started an internship at Pazhou Laboratory, working on SFT post-training for music style transfer.

šŸ“Selected Publications

ICML 2026 ICML 2026 poster for Steer Where It Matters

Steer Where It Matters: Token-Level Visual-Sensitivity Steering for LVLMs Hallucination Mitigation

Ruipeng Zhang, Zhihao Li, C. L. Philip Chen, Tong Zhang

This work proposes a lightweight token-level visual-sensitivity steering method for hallucination mitigation in large vision-language models.

ICLR 2026 ICLR 2026 poster for P²-DPO

P2-DPO: Grounding Hallucination in Perceptual Processing via Calibration Direct Preference Optimization

Ruipeng Zhang, Zhihao Li, Haozhang Yuan, C. L. Philip Chen, Tong Zhang

This work studies hallucination mitigation from the perspective of perceptual processing and preference optimization for large vision-language models.

šŸŽ“Educations

South China University of Technology

M.Eng. in Computer Technology

Advisor: Prof. Tong Zhang

Jinan University

B.Eng. in Computer Science and Technology

šŸ† GPA: 4.09/5.00; ranked 3/99; National Scholarship recipient

šŸ’¼Internships

2026.05 - Present

ByteDance, Douyin Department

Research Intern

I work on agentic reinforcement learning, vision-to-code, and long-horizon agent tasks.

2025.08 - 2026.04

China Southern Power Grid AI Company

Research Intern

Computer use agents.

2025.03 - 2025.07

Pazhou Laboratory

Research Intern

SFT for music style transfer.

Reviewer

  • ICLR 2027
  • AAAI 2027