GD^2PO: Mitigating Multi-Reward Conflicts via Group-Dynamic reward-Decoupled Policy Optimization Paper • 2606.16771 • Published 23 days ago • 13
LoopCoder-v2: Only Loop Once for Efficient Test-Time Computation Scaling Paper • 2606.18023 • Published 22 days ago • 209
Embodied-R1.5: Evolving Physical Intelligence via Embodied Foundation Models Paper • 2606.11324 • Published 29 days ago • 170
EvalVerse: Pipeline-Aware and Expert-Calibrated Benchmarking for Professional Cinematic Video Generation Paper • 2605.23271 • Published May 22 • 81
ThoughtTrace: Understanding User Thoughts in Real-World LLM Interactions Paper • 2605.20087 • Published May 19 • 18
Anti-Self-Distillation for Reasoning RL via Pointwise Mutual Information Paper • 2605.11609 • Published May 12 • 196
FlowLong: Inference-time Long Video Generation via Manifold-constrained Tweedie Matching Paper • 2605.20910 • Published May 20 • 29