Flash-GRPO: Efficient Alignment for Video Diffusion via One-Step Policy Optimization
Paper • 2605.15980 • Published • 36
Reinforcement learning
-
-
NGRPO: Negative-enhanced Group Relative Policy Optimization
Paper • 2509.18851 • Published • 2 -
CEPO: RLVR Self-Distillation using Contrastive Evidence Policy Optimization
Paper • 2605.19436 • Published • 14 -
Delta Attention Residuals
Paper • 2605.18855 • Published • 8 -
Steered LLM Activations are Non-Surjective
Paper • 2604.09839 • Published • 13 -
GQLA: Group-Query Latent Attention for Hardware-Adaptive Large Language Model Decoding
Paper • 2605.15250 • Published • 13 -
Self-Distilled Agentic Reinforcement Learning
Paper • 2605.15155 • Published • 113 -
AnyFlow: Any-Step Video Diffusion Model with On-Policy Flow Map Distillation
Paper • 2605.13724 • Published • 102 -
F-GRPO: Factorized Group-Relative Policy Optimization for Unified Candidate Generation and Ranking
Paper • 2605.12995 • Published • 2 -
World Action Models: The Next Frontier in Embodied AI
Paper • 2605.12090 • Published • 67 -
AlphaGRPO: Unlocking Self-Reflective Multimodal Generation in UMMs via Decompositional Verifiable Reward
Paper • 2605.12495 • Published • 35 -
World Model for Robot Learning: A Comprehensive Survey
Paper • 2605.00080 • Published • 16 -
Flow-OPD: On-Policy Distillation for Flow Matching Models
Paper • 2605.08063 • Published • 101 -
Learning beyond Teacher: Generalized On-Policy Distillation with Reward Extrapolation
Paper • 2602.12125 • Published • 67 -
KEPO: Knowledge-Enhanced Preference Optimization for Reinforcement Learning with Reasoning
Paper • 2602.00400 • Published -
OVD: On-policy Verbal Distillation
Paper • 2601.21968 • Published • 6 -
SODA: Semi On-Policy Black-Box Distillation for Large Language Models
Paper • 2604.03873 • Published • 2 -
Post-Trained MoE Can Skip Half Experts via Self-Distillation
Paper • 2605.18643 • Published • 30