Guowei Xu
Scholar

Guowei Xu

Google Scholar ID: 7xKdIM4AAAAJ
Tsinghua University
Language ModelsReinforcement Learning
Citations & Impact
All-time
Citations
329
 
H-index
5
 
i10-index
4
 
Publications
7
 
Co-authors
12
list available
Resume (English only)
Academic Achievements
  • Published multiple papers at top-tier conferences (NeurIPS, ICCV, ICML, ICLR) as first or co-first author
  • Paper 'ACE: Off-Policy Actor-Critic with Causality-Aware Entropy Regularization' accepted as Oral at ICML 2024
  • Paper 'DrM: Mastering Visual Reinforcement Learning through Dormant Ratio Minimization' accepted as Spotlight at ICLR 2024
  • Official implementation of LLaVA-CoT (GitHub ⭐2087)
  • Official implementation of DrM (GitHub ⭐76)
  • Reviewer for ICLR (2025, 2026) and ICML (2025)
  • Overall winner (best total score) at the 52nd International Physics Olympiad (IPhO 2022)
  • National Scholarship recipient in 2024 and 2025
  • Selected for Tsinghua University’s Sparking Program in 2025 (Top 1%)
  • First Prize Freshman Scholarship, Tsinghua University, 2024