6 papers
Prorsa: Unleashing the Potential of Prefill-Decode Disaggregation in Agentic Reinforcement Learning
Chiheng Lou, Xiangwei Deng, Bangjun Xiao +4
Agentic reinforcement learning (RL) is commonly used to enhance the capabilities of large language models (LLMs) in solving complex multi-turn tasks. During rollout, agentic RL exh…
CodeMidas: Scaling Agentic Coding RL Environments from Code Itself
Bowen Ye, Lei Li, Shicheng Li +16
Training capable coding agents via reinforcement learning (RL) requires diverse tasks with reliable verifiers. Open-source codebases offer a rich source of such tasks, while existi…
ARL-Tangram: Unleash the Resource Efficiency in Agentic Reinforcement Learning
Bangjun Xiao, Yihao Zhao, Xiangwei Deng +9
Agentic reinforcement learning (RL) has emerged as a transformative workload in cloud clusters, enabling large language models (LLMs) to solve complex problems through interactions…
HySparse: A Hybrid Sparse Attention Architecture with Oracle Token Selection and KV Cache Sharing
Yizhao Gao, Jianyu Wei, Qihao Zhang +11
This work introduces Hybrid Sparse Attention (HySparse), a new architecture that interleaves each full attention layer with several sparse attention layers. While conceptually simp…
Reinforcement Learning for Chain of Thought Compression with One-Domain-to-All Generalization
Hanyu Li, Jiangshan Duo, Bofei Gao +4
Chain-of-thought reasoning in large language models can trigger an "overthinking trap": longer rollouts raise cost and latency yet often yield unreliable accuracy gains. Existing m…
JudgeRLVR: Judge First, Generate Second for Efficient Reasoning
Jiangshan Duo, Hanyu Li, Hailin Zhang +3
Reinforcement Learning with Verifiable Rewards (RLVR) has become a standard paradigm for reasoning in Large Language Models. However, optimizing solely for final-answer correctness…