arXiv AI

ROSS: Relearning from Self-Generated Rollouts through Selective Supervision

arXiv Machine Learning
Sep 14

MInTRL: Off-policy Intervention can boost On-policy RL

MInTRL (Minimal Intervention Reinforcement Learning) expands exploration in on-policy reinforcement learning by inserting sparse, local corrections into rollouts via a judge-intervention policy. These interventions replace erroneous suffixes and immediately return control to the main policy, allowing the agent to explore beyond its natural trajectory while maintaining on-policy data. The method uses a sequence-level advantage-regression objective, avoiding importance sampling, and demonstrates superior performance on math and code benchmarks compared to standard on-policy and off-policy baselines.

By Mingyu Chen, Yefan Tao, Gerald Friedland, Xuezhou Zhang, Chris Kong
arXiv Machine Learning
Oct 8

RollVerify: Bridging Efficiency and Accuracy in Long-Tail Rollout Reinforcement Learning

RollVerify is a lightweight reinforcement learning framework that addresses the trade‑off between efficiency and accuracy in long‑tail rollout settings. It introduces an off‑policy shift metric (OPS) to quantify deviation in partially generated trajectories and uses sequence‑level and token‑level verification to truncate invalid suffixes before training. Experiments on mathematical reasoning tasks show that RollVerify matches on‑policy accuracy while cutting training costs, with preliminary evidence from code‑generation tasks.

By Yongqiang Yao, Jinru Tan, Kaihuan Liang, Zixin Yin, Yazhe Niu, Ruihao Gong, Dahua Lin, Ningyi Xu
Hugging Face Trending Papers
Jul 21

REGEN: Replay-recycling for Expert-to-Generalist distillation with Offline Reinforcement Learning

Large-scale online reinforcement learning (RL) is the predominant means of eliciting advanced abilities including long-term reasoning and agentic tool use in large language models (LLMs). However, continuing to scale it across vast task domains of interest remains challenging in both computational infrastructure and cost, especially when considering RL as merely a one-off learning stage.

arXiv Machine Learning
Oct 12

Cross-Behavior Policy Optimization: Unified Learning from Offline Demonstrations and On- and Off-Policy Rollouts via a Shared Regression Target

The paper introduces Cross-Behavior Policy Optimization (CBPO), a method that unifies offline demonstrations and policy-generated rollouts into a single regression loss using a fixed-reference target. CBPO eliminates the need for separate supervised fine‑tuning and reinforcement learning objectives, allowing positive and negative offline completions, reference responses, and current and historical rollouts to share a common reward‑labeled format and replay buffer. Experiments on two model scales and four mathematical reasoning benchmarks show that CBPO, especially the CBPO‑Unified variant, outperforms baseline methods in avg@32 and pass@32 metrics, with historical rollouts providing additional gains.

By Du Cheng, Junyu Lu, Jingfan Zhang, Yuzhuo Zhang, Haoxuan Jia, Shichao Weng
Hugging Face Trending Papers
Oct 5

Off-Policy Merging Beats On-Policy Self-Distillation for Continual Learning

The paper demonstrates that off‑policy merging, called grafting, outperforms on‑policy self‑distillation for continual learning. Grafting learns updates on an earlier donor checkpoint, scales the update, and optionally masks sensitive directions, thereby reducing interference with existing capabilities. Across various continual learning scenarios, grafting achieves better new‑task and old‑task performance without costly on‑policy sampling.

arXiv Computation and Language
Oct 28

CAMR: Compatibility-Aware Memory Replay for Continual Learning in Large Language Models

CAMR (Compatibility-Aware Memory Replay) is an adaptive framework for continual learning in large language models that jointly manages supervision of new examples and replay of historical data. It uses compatibility checks and a learned memory gate to decide among three actions: general replay of incoming examples, reinforcing compatible related examples, and skipping supervision for mastered duplicates. Evaluated on Qwen3-8B-Base, CAMR outperforms existing methods on WikiBigEdit factual updates and achieves top scores on multiple TRACE tasks while preserving earlier-task performance across diverse domains.

By Jingpu Yang, Tianhao Li, Fengxian Ji, Yufeng Wang, Shuxin Zheng, Xiaoqing Zhang