离线强化学习遇上LLM推理，未来策略近似方法带来新突破，AI研究者不容错过。
arXiv:2509.19893v3 Announce Type: replace Abstract: Reinforcement learning (RL) has emerged as a key driver of post-training for complex reasoning in …
提出分层动作分块机制，让离线RL能够高效学习长程动作序列，显著提升复杂任务成功率。
arXiv:2607.20834v1 Announce Type: new Abstract: Offline goal-conditioned reinforcement learning (RL) holds the promise of learning general-purpose pol…
离线强化学习如何驾驭LLM智能体？这篇论文提出了新方法，值得AI研究者关注。
arXiv:2607.05458v1 Announce Type: cross Abstract: Large language model (LLM) agents are usually improved by changing prompts, models, or hand-written …
离线强化学习新范式：后验混合贝叶斯信念正则化策略优化，破解分布外动作评估难题
arXiv:2606.00680v1 Announce Type: cross Abstract: Offline reinforcement learning (RL) aims to optimize policies from pre-collected datasets. A bottlen…
用离线强化学习高效微调大模型，代码生成性能与训练成本双优化
arXiv:2605.28409v1 Announce Type: new Abstract: Post-training using online reinforcement learning (RL) is an important training step for LLMs, includi…
用离线强化学习实现全局预算感知的LLM顺序路由，解决多模型调用成本与性能平衡难题。
arXiv:2605.25424v1 Announce Type: cross Abstract: Existing LLM routing frameworks treat queries as independent events, neglecting the sequential natur…
跨领域离线强化学习新方法，用对齐贝尔曼备份解决域间转移难题。
arXiv:2605.22376v1 Announce Type: new Abstract: Cross-domain offline reinforcement learning (CDRL) aims to improve policy learning in a target domain …
提出通用视界模型，直接预测折扣无限期未来，缓解离线RL模型推断的复合误差
arXiv:2605.15603v1 Announce Type: cross Abstract: Model-based reinforcement learning (RL) offers a compelling approach to offline RL by enabling value…