AV-GRPO: Modality-Anchored Decoupling Diffusion Reinforcement Learning for Joint Audio-Video Generation
WanPE: Towards Cinematic Prompt Enhancement for Modern Text-to-Video Generation
When and Where to Trust the Teacher: Unifying On-Policy Distillation and GRPO through Entropy-Calibrated Credit Assignment
SpeakerMem-R1: Speaker-Centered Dual-Track Memory for Multi-Party Dialogue
XiaomiMiMo/MiMo-V2.6-Flash-RL — new model trending #30 on Hugging Face
CodeMidas: Scaling Agentic Coding RL Environments from Code Itself
1
$λ$-Controlled GRPO: Turning Flow-Matching Ratio Instability into a Budgeted Resource
CodeMidas: Scaling Agentic Coding RL Environments from Code Itself
2
Don't Mask the Environment: Observation Supervision Changes How Agents Explore Under RL
Multi-Dimensional Prosody Judgment For Live Streaming Speech Synthesis
1
Google Research Introduces Retrieve-for-Train (R4T): An RL-Compiled Diffusion Retriever for 12× to 20× Faster Query Fan-Out
4 min1
Don't Mask the Environment: Observation Supervision Changes How Agents Explore Under RL
Not All Prompts Are Equal: Exploration-Guided Prompt Scaffolding for Multimodal Reinforcement Post-Training
Expert-Space Exploration in MoE Reinforcement Learning
Learning to Solve Hard Problems in RL for LLMs by Never Giving Up
Expert-Space Exploration in MoE Reinforcement Learning
Async GRPO with LoRA across HF Jobs: a bucket, a proxy, and no NCCL
ActReview: Rebuttal-Guided Training Data and Rubric Rewards for Actionable Peer Review Generation
1
ThinkPrior: Zero-Rollout Difficulty Priors for Cold-Start Prompt Selection in RLVR
1
ActReview: Rebuttal-Guided Training Data and Rubric Rewards for Actionable Peer Review Generation
DataFlex-RL: An Evaluation Platform for RLVR Data Policies
Multi-Step Tool-Calling over Korean Open Public APIs: A Benchmark and a Data-Synthesis Recipe
1
Fine-tuning a 350M Model for Better Structured Outputs in 100 GRPO Steps
Verify Before You Distill: Prompt-Level Teacher Gating for On-Policy Distillation
