← 返回概念图谱 引用本概念的论文(16)
- DeepSearch-World：可验证环境中面向深度搜索智能体的自蒸馏方法 DeepSearch-World: Self-Distillation for Deep Search Agents in a Verifiable Environment arXiv 2607.07820
- UCOB：通过信用感知的在线策略双向自蒸馏学习利用与演化智能体技能 UCOB: Learning to Utilize and Evolve Agentic Skills via Credit-Aware On-Policy Bidirectional Self-Distillation arXiv 2606.29502
- 为何带反馈增强的自蒸馏无法提升检索交错搜索代理？ Why Does Feedback-Augmented Self-Distillation Fail to Improve Retrieval-Interleaved Search Agents? arXiv 2607.17558
- 在策略自蒸馏中诊断与缓解思维坍缩 Diagnosing and Mitigating Thinking Collapse in On-Policy Self-Distillation arXiv 2607.10805
- 将共识作为特权上下文用于无标签自蒸馏 Consensus as Privileged Context for Label-Free Self-Distillation arXiv 2607.13643
- 通过自蒸馏增强基于评分标准的强化学习 Enhancing Rubric-based RL via Self-Distillation arXiv 2607.18082
- 面向免标注大语言模型自蒸馏的神经元感知数据选择 Neuron-Aware Data Selection for Annotation-Free LLM Self-Distillation arXiv 2607.02460
- 将共识作为特权上下文用于无标签自蒸馏 Consensus as Privileged Context for Label-Free Self-Distillation arXiv 2607.13643
- 更好的起点，更好的终点：用于压缩推理的自举迭代自推理蒸馏 Better Starts, Better Ends: Bootstrapped Iterative Self-Reasoning Distillation for Compressed Reasoning arXiv 2607.15736
- 重新思考面向思维模型的在策略自蒸馏 Rethinking On-Policy Self-Distillation for Thinking Models arXiv 2607.05184
- DRIFT：基于难度路由自蒸馏、节奏门控探索与成功缓冲区训练的方法 DRIFT: Difficulty Routing Self-DIstillation with Rhythm-Gated Exploration and Success BuFfer Training arXiv 2606.30345
- KbSD：面向智能体搜索中行为校准的知识边界感知自蒸馏 KbSD: Knowledge Boundary aware Self-Distillation for Behavioral Calibration in Agentic Search arXiv 2606.29863
- SEED：面向智能体强化学习的自演化在策略蒸馏方法 SEED: Self-Evolving On-Policy Distillation for Agentic Reinforcement Learning arXiv 2607.14777
- 诊断与缓解在线策略自蒸馏中的思维坍缩 Diagnosing and Mitigating Thinking Collapse in On-Policy Self-Distillation arXiv 2607.10805
- EasyOPD：面向大语言模型的易用式在策略蒸馏框架 EasyOPD: An Easy-to-use On-Policy Distillation Framework for Large Language Models arXiv 2607.11012
- BucketKD：一种面向端到端运动规划的安全感知分桶知识蒸馏框架 BucketKD: A Safety-Aware Bucket-Based Knowledge Distillation Framework for End-to-End Motion Planning arXiv 2607.10565
共现相关概念
在引用本概念的论文中,与下列概念同时出现的次数(降序)。
