---

• GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization

  Paper review • #RL #alignment #reward weight

  January, 2026

• Reinforcement Pre-Training

  Paper review • #RL #pre-training

  January, 2026

• SimMMDG: A Simple and Effective Framework for Multi-modal Domain Generalization

  Paper review • #OOD #multimodal learning

  October, 2025

• ACL conference note

  ACL'24 paper list and repo

  August, 2024