合成语音训练资源稀缺？本文提出GRPO方法，突破仅用合成语音的局限，为语音AI带来新思路
arXiv:2607.08409v1 Announce Type: new Abstract: LLM-based ASR adapted to regulated domains such as banking is bottlenecked by privacy: real speech is …
大模型CoT推理中置信度与理由对齐的新方法，GRPO优化让推理更可靠。
arXiv:2606.14961v1 Announce Type: new Abstract: Chain-of-thought (CoT) reasoning can improve LLM performance, but high answer confidence may be mislea…
用强化学习优化大模型事件预测，GRPO策略带来新突破
arXiv:2606.15917v1 Announce Type: new Abstract: We use Group Relative Policy Optimization (GRPO), a recently devised sample and memory efficient reinf…
用方差感知评分奖励与GRPO方法提升大模型心脏医学问答准确性，为医疗AI提供新思路。
arXiv:2606.05174v1 Announce Type: cross Abstract: Large Language Models (LLMs) have shown strong promise in healthcare applications. Yet deploying gen…
听觉大模型新突破，FSA-GRPO方法教会模型巧用少样本演示，大幅提升推理与适应能力。
arXiv:2606.02615v1 Announce Type: cross Abstract: Few-shot prompting provides an effective way to adapt auditory large language models to low-resource…
小模型天然具备策略级探索优势，在GRPO中提升多样性比传统token随机更稳定高效
arXiv:2605.30789v1 Announce Type: cross Abstract: We identify a new dimension for enhancing rollout diversity in Group Relative Policy Optimization (G…
英伟达开源Polar框架，让Codex等智能体通过GRPO训练后性能暴涨594.74%，代码AI能力再上新台阶。
IT之家 5 月 28 日消息，英伟达研究团队本周发布开源框架 Polar，在不破坏原有工具调用、上下文组织和补丁提交方式的前提下，让 Codex、Claude Code、Qwen Code 等现有智能体框架接入 GRPO（广义相对策略优化）训练。 IT之家注：GRPO 是一种面向强化学习训练的优化…
通过pass-rate加权自蒸馏，恢复LLM推理的“甜蜜点”，破解GRPO归一化带来的学习偏差。
arXiv:2605.27765v1 Announce Type: cross Abstract: Self-Distillation Policy Optimization (SDPO) provides dense token-level credit assignment for reinfo…
提出IRPO方法，利用后训练GRPO显著提升图像恢复性能，为低层视觉任务带来新突破。
arXiv:2512.00814v3 Announce Type: replace Abstract: Post-training has become effective for high-level generation, but its role in low-level vision rem…
强化学习新突破：XRPO通过定向探索与利用，显著提升GRPO在LLM推理任务中的性能与稳定性
arXiv:2510.06672v3 Announce Type: replace Abstract: Reinforcement learning algorithms such as GRPO have driven recent advances in large language model…
用GRPO优化技术将推理链转化为代码，有效提升低资源语言的代码生成能力，ICLP 2026收录的硬核研究。
arXiv:2506.11027v3 Announce Type: replace-cross Abstract: Generating accurate and executable code using Large Language Models (LLMs) remains a signifi…
GRPO新变体F-TIS：通过多模型协作提升LLM后训练奖励信号多样性，突破单一策略局限。
arXiv:2605.22537v1 Announce Type: new Abstract: Reinforcement learning methods such as GRPO have seen great popularity in LLM post-training. In GRPO, …
让保守的LLM在任务对话中主动出击，论文用奖励塑造RL解锁销售场景的前瞻策略。
arXiv:2605.22240v1 Announce Type: new Abstract: Proactive task-oriented dialogue (TOD), such as outbound sales, demands a persuasive agent that active…
从价值梯度假说揭秘PPO和GRPO为何有效，为LLM后训练提供新理论框架。
arXiv:2605.21654v1 Announce Type: cross Abstract: Reinforcement learning substantially improves pretrained language models, but it remains understudie…
一篇系统梳理LLM后训练中强化学习的综述，涵盖RLHF、DPO、RLVR等前沿方法
arXiv:2407.16216v4 Announce Type: replace Abstract: Large language models (LLMs) trained via pretraining and supervised fine-tuning (SFT) can still pr…
提出稀疏到稠密奖励原则，四阶段后训练流程更高效利用稀缺标注数据，为LLM推理优化提供新范式。
arXiv:2605.12483v2 Announce Type: replace-cross Abstract: When labeled verifiable training data is scarce, each checked example should be used where i…
将工具调用与执行解耦，提出隐式层次化GRPO框架，显著提升数学推理中的工具集成效率与泛化能力。
arXiv:2605.18500v1 Announce Type: new Abstract: Large language models (LLMs) have increasingly leveraged tool invocation to enhance their reasoning ca…
最新研究揭示GRPO方法在线策略局限，Mu-GRPO通过离线策略提升LLM强化学习效率，降低计算成本。
arXiv:2605.17570v1 Announce Type: new Abstract: Group Relative Policy Optimization (GRPO) has been a key driver of recent progress in reinforcement le…
提出概率块掩码机制，直击VLA强化学习后训练计算瓶颈，显著提升效率。
arXiv:2605.16154v1 Announce Type: new Abstract: Reinforcement learning (RL) allows vision-language-action (VLA) policies to generalize beyond their tr…
最新研究揭示重要性采样在结果监督RL中的信用分配偏差，提出不对称比率改进方法，为LLM后训练提供新视角。
arXiv:2510.06062v2 Announce Type: replace Abstract: Reinforcement learning (RL) has shown great promise in large language models (LLMs) post-training,…