#policy-optimization
- DRPO: Rethinking Divergence Regularization in LLM Reinforcement Learning Tencent Hunyuan research
- Flow-DPPO: Principled RL Alignment for Flow Matching Image and Video Models Tencent Hunyuan research
- Beyond Euclidean Clipping: Overcoming Exploration Collapse in LLM RL via Riemannian Isometric Policy Optimization research
- Rethinking Critic Learning in PPO: Understanding and Mitigating Value Flattening Shanghai AI Laboratory research
- ZPPO: Teacher-in-Prompts Knowledge Distillation Outperforms Gradient Methods for Small Reasoners NVIDIA research