cs.AI（2026-06-10）

📊 共 17 篇论文 | 🔗 5 篇有代码

🎯 兴趣领域导航

支柱二：RL算法与架构 (RL & Architecture) (9 🔗3) 支柱九：具身大模型 (Embodied Foundation Models) (8 🔗2)

🔬 支柱二：RL算法与架构 (RL & Architecture) (9 篇)

#	题目	一句话要点	标签	🔗	⭐
1	MODF-SIR: A Multi-agent Omni-modal Distilled Framework for Social Intelligence Reasoning	提出多代理全模态蒸馏框架以解决社会智能推理问题	distillation large language model foundation model	✅
2	Fast Speech Foundation Model Distillation Using Interleaved Stacking	提出交错堆叠方法加速语音基础模型蒸馏	distillation foundation model
3	SVoT: State-aware Visualization-of-Thought for Spatial Reasoning via Reinforcement Learning	提出SVoT以解决多模态大语言模型的空间推理问题	reinforcement learning reward design large language model
4	Frozen Multimodal Embeddings for Personality and Cognitive Ability Assessment in Asynchronous Video Interviews	提出冻结多模态嵌入以评估视频面试中的个性与认知能力	representation learning multimodal
5	Lung-SRAD: Spectral-Aware Regularized Audio DASS with Dual-Axis Patch-Mix Contrastive Learning for Respiratory Sound Classification	提出Lung-SRAD以解决呼吸声音分类中的局部异常检测问题	SSM state space model representation learning	✅
6	Architecture-Aware Reinforcement Learning Makes Sliding-Window Attention Competitive in Math Reasoning	提出SWARR以解决长上下文推理中的自注意力效率问题	reinforcement learning large language model
7	HERO: Hindsight-Enhanced Reflection from Environment Observations for Agentic Self-Distillation	提出HERO框架以解决多回合强化学习中的反馈对齐问题	reinforcement learning distillation
8	CCKS: Consensus-based Communication and Knowledge Sharing	提出CCKS框架以解决多智能体强化学习中的知识共享问题	reinforcement learning contrastive learning teacher-student	✅
9	Organize then Retrieve: Hierarchical Memory Navigation for Efficient Agents	提出HORMA以解决长任务中的记忆效率问题	reinforcement learning large language model

🔬 支柱九：具身大模型 (Embodied Foundation Models) (8 篇)

#	题目	一句话要点	标签	🔗	⭐
10	Towards Data-free and Training-free Compression for Speech Foundation Models Using Parameter Clustering	提出无数据和无训练的语音基础模型压缩方法以解决模型体积问题	foundation model
11	A Lightweight Multi-Agent Framework for Automated Concrete Barrier Design	提出轻量级多智能体框架以实现自动化混凝土障碍物设计	large language model	✅
12	TAHOE: Text-to-SQL with Automated Hint Optimization from Experience	提出TAHOE以解决文本到SQL转换中的提示优化问题	large language model
13	Nonslop: A Gamified Experiment in Human-AI Collaborative Writing	通过游戏化写作实验探讨人类与AI的协作创作	large language model
14	Rule Taxonomy and Evolution in AI IDEs: A Mining and Survey Study	提出AI IDE规则分类与演变研究以优化开发者体验	large language model
15	Grammar-Constrained Decoding Can Jailbreak LLMs into Generating Malicious Code	提出CodeShield以解决GCD引发的LLM恶意代码生成问题	large language model
16	Can Open-Source LLM Agents Replace Static Application Security Testing Tools? An Empirical Assessment	评估开源LLM代理在静态应用安全测试中的有效性	large language model
17	SkillJuror: Measuring How Agent Skill Organization Changes Runtime Behavior	提出SkillJuror框架以评估智能体技能组织对行为的影响	large language model	✅

⬅️ 返回 cs.AI 首页 · 🏠 返回主页