The paper addresses the memory bottleneck in reinforcement learning for large language models caused by the large Key-Value (KV) cache during rollout phases. It highlights that while KV cache compression can reduce memory usage, it introduces a significant off‑policy bias that standard statistical corrections cannot adequately mitigate. The authors argue that even tiny compression errors are amplified by RL’s instability, leading to inefficient learning.
By Rui Zhu, Weiheng Bai, Qiushi Wu, Yang Ren, Haixu Tang, Yuchu Liu
arXiv:2608. 01418v1 Announce Type: cross Abstract: Autoregressive rollout generation is a major computational cost in reinforcement learning for large language models.
By Wenhao Zhang, Yibo Xie, Rui Wang, Jiahua Yang, Lei Jiang, Zibo Yang, Yawei Wang, Jiali Xu, jasperawang, Haoyang Long, Huan Xiong, alantzhao
The paper proposes CANOPY, a minimalist reinforcement learning protocol that addresses two common pitfalls—signal starvation and policy drift—in outcome‑only RL for long‑horizon interactive tasks. By scaling same‑task exploration, keeping updates on‑policy, and anchoring updates with KL divergence, CANOPY enables a Qwen3‑14B agent to achieve top leaderboard results on the AppWorld coding benchmark without auxiliary supervision or elaborate scaffolding. The approach also improves performance on SWE‑bench for a Qwen3.5‑9B model.
By Liming Pu, Xiaoxia Li, Yifu Liu, Teng Cao, Bin Yang
arXiv:2607. 05394v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) is a powerful recipe for improving language-model reasoning, but it is expensive to repeat on every new strong model because the target model must generate many rollouts during training.
By Shiyuan Feng, Huan-ang Gao, Haohan Chi, Hanlin Wu, Zhilong Zhang, Zheng Jiang, Bingxiang He, Wei-Ying Ma, Ya-Qin Zhang, Hao Zhou
Reinforcement learning (RL) has become a key component in modern large language models, yet the rollout stage remains the key bottleneck in RL training pipelines. Although Multi-Token Prediction (MTP) offers a natural solution to accelerate rollouts through speculative decoding, many studies have observed that MTP acceptance rates degrade significantly during RL training, leading to limited speedup performance.
arXiv:2607.08837v4 Announce Type: replace-cross
Abstract: Exploration is essential to RL since a policy cannot improve by repeatedly sampling the behaviors it already prefers. Standard methods inject...
By Sunshine Jiang, John Marangola, David Zhang, Raghuram Kowdeed, Ruiyang Luo, Nitish Dashora, Richard Li, Pulkit Agrawal, Zhang-Wei Hong
Reinforcement learning with verifiable rewards (RLVR) is a powerful recipe for improving language-model reasoning, but it is expensive to repeat on every new strong model because the target model must generate many rollouts during training. As models scale, post-training itself becomes a bottleneck.
arXiv:2607. 16204v1 Announce Type: new Abstract: Recent growth in reinforcement learning (RL) has surfaced a need for diverse, specialized training environments.
By Darshan Deshpande
arXiv:2606. 12370v1 Announce Type: new Abstract: Reinforcement learning (RL) has become a key component in modern large language models, yet the rollout stage remains the key bottleneck in RL training pipelines.
By Yucheng Li, Huiqiang Jiang, Yang Xu, Jianxin Yang, Yi Zhang, Yizhong Cao, Yuhao Shen, Fan Zhou, Rui Men, Jianwei Zhang, An Yang, Bowen Yu, Bo Zheng, Fei Huang, Junyang Lin, Dayiheng Liu, Jingren Zhou
Reinforcement learning (RL) is becoming increasingly important for post-training large language models (LLMs). Previous RL pipelines for LLMs were mostly synchronous and batch-interleaved, which is inefficient for long-horizon agentic tasks.
arXiv:2602. 04879v3 Announce Type: replace-cross Abstract: Reinforcement learning (RL) has become a cornerstone for fine-tuning Large Language Models (LLMs), with Proximal Policy Optimization (PPO) serving as the de facto standard algorithm.
By Penghui Qi, Xiangxin Zhou, Zichen Liu, Tianyu Pang, Chao Du, Min Lin, Wee Sun Lee
GrowMTP is a method that trains a draft head entirely within the reinforcement learning (RL) loop, using supervision from the RL verification step and a rollout distribution that is narrower than pretraining. By detaching draft‑head updates from the policy backbone, it enables online training of the draft head from scratch. Experiments on Qwen3‑4B, MiMo‑7B‑SFT, and Qwen3.5‑4B‑Base show rollout speedups ranging from 1.36× to 2.13× and overall end‑to‑end speedups from 1.20× to 1.60×, making it a modular acceleration component for RL frameworks lacking pretrained draft heads.
By Minghua He, Lingzhe Zhang, Yuan Liu, Xiao Zhou, Aiwei Liu