The paper introduces a Markov chain Monte Carlo (MCMC) sampling algorithm that transforms off‑policy traces into more on‑policy data for supervised finetuning (SFT). By tailoring the data distribution rather than changing the learning objective, the method allows SFT to match or surpass traditional post‑training techniques such as reinforcement learning across tasks like scientific skill acquisition, mathematical reasoning, and open‑ended expertise. The resulting models generalize better, forget less, and exhibit strong distributional performance, demonstrating that sampling can serve as a model‑native operator for broader post‑training applications.
By Aayush Karan, Sitan Chen, Yilun Du
The paper investigates how reusing past samples can improve the sample efficiency of Proximal Policy Optimization (PPO). Two variants, wPPO-U and wPPO-BH, are introduced within a multiple importance weighting framework, each reusing data from recent iterations while preserving core PPO mechanics. The authors derive theoretical policy improvement bounds for both variants and empirically evaluate their impact on continuous control tasks.
By Alessandro Montenegro, Riccardo Venturelli, Marco Mussi, Matteo Papini, Alberto Maria Metelli
Reward models (RMs) provide critical feedback signals for LLM post-training, notably in reinforced fine-tuning (RFT) and reinforcement learning (RL) pipelines. However, current reward evaluation relies on heterogeneous criteria such as rule-based verifiers, ground-truth references, procedural checklists, and complex rubrics, where a unified mechanism to integrate all types of evidence remains unexplored.
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:2607. 04728v1 Announce Type: cross Abstract: Reinforcement learning (RL) post-training for large language models (LLMs) follows a efficient paradigm of "rollout then update", which inevitably results in off-policy training data.
By Yu Li, Xiuyu Li, Mingyang Yi, Jiaxing Wang, zhangliangxu, Zhaolong Xing, Zhen Chen
arXiv:2607. 11720v1 Announce Type: cross Abstract: Background: Offline reinforcement learning (RL) enables effective policies to be trained from large, previously collected datasets and subsequently improved through limited online interaction.
By Alper Kamil Bozkurt, Shangtong Zhang, Yuichi Motai
arXiv:2606. 01281v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) has emerged as a powerful paradigm for enhancing the reasoning capabilities of large language models (LLMs).
By Yixiu Mao, Yun Qu, Qi Wang, Heming Zou, Xiangyang Ji
The paper addresses instability in off‑policy reinforcement learning for large language models (LLMs) caused by large policy lag. It shows that the correct correction for the prefix‑marginal policy gradient is the cumulative prefix importance ratio, which token‑level weighting fails to capture. To improve stability, the authors propose Minimum Prefix Ratio (MinPRO), a surrogate objective that uses the minimum token‑level ratio in the preceding prefix, and demonstrate its effectiveness on dense and mixture‑of‑experts LLMs across mathematical reasoning benchmarks.
By Shiye Lei, Zhihao Cheng, Dacheng Tao
arXiv:2610.19783v1 Announce Type: new
Abstract: Reinforcement learning (RL) holds the promise of learning a policy that can outperform what is shown in the data. However, in practice, such systems su...
By Zhiyuan Zhou, Sergey Levine
arXiv:2602. 21492v2 Announce Type: replace-cross Abstract: Reinforcement learning (RL) has become a central post-training paradigm for large language models (LLMs), but its performance is highly sensitive to the quality of training problems.
By Ningyuan Yang, Weihua Du, Weiwei Sun, Sean Welleck, Yiming Yang
Reinforcement learning (RL) post-training for large language models (LLMs) follows a efficient paradigm of "rollout then update", which inevitably results in off-policy training data. To resolve this, Importance sampling (IS) is proposed, while the token-level ratios compound over long sequences, causing severe variance exploded.
arXiv:2609.00213v1 Announce Type: new
Abstract: Reinforcement learning fine-tuning of large language models increasingly adopts multiple reward dimensions, including verifiable rules, task-specific e...
By Yu Yuan, Yaoyou Fan, Lili Zhao, Guangting Zheng, Kai Zhang, Lu Pan, Ke Zeng, Qi Liu