cs.AI（2026-05-06）

📊 共 20 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (15 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (5)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (15 篇)

#	题目	一句话要点	标签	🔗	⭐
1	A Foundation Model for Zero-Shot Logical Rule Induction	提出神经规则归纳器NRI，用于零样本逻辑规则归纳的基础模型。	foundation model zero-shot transfer	✅
2	Position: Embodied AI Requires a Privacy-Utility Trade-off	提出SPINE框架，解决具身智能在敏感环境中部署时的隐私-效用权衡问题	embodied AI	✅
3	SoK: Robustness in Large Language Models against Jailbreak Attacks	提出Security Cube框架，系统评估大语言模型抵抗越狱攻击的鲁棒性	large language model
4	JASTIN: Aligning LLMs for Zero-Shot Audio and Speech Evaluation via Natural Language Instructions	JASTIN：通过自然语言指令对齐LLM，实现零样本音频和语音评估	large language model multimodal instruction following
5	From Parameter Dynamics to Risk Scoring : Quantifying Sample-Level Safety Degradation in LLM Fine-tuning	提出SQSD方法，量化样本级安全退化风险，提升LLM微调安全性	large language model
6	Deployment-Relevant Alignment Cannot Be Inferred from Model-Level Evaluation Alone	模型层面对齐评估无法推断部署相关的对齐，需系统级评估	instruction following
7	Almost-Orthogonality in Lp Spaces: A Case Study with Grok	提出改进的三函数不等式以解决Lp空间中的几何问题	large language model
8	Think-Aloud Reshapes Automated Cognitive Model Discovery Beyond Behavior	利用“思考发声”重塑自动认知模型发现，超越行为数据本身	large language model
9	Uno-Orchestra: Parsimonious Agent Routing via Selective Delegation	Uno-Orchestra：通过选择性委托实现精简的多智能体路由	large language model
10	AISSA: Implementation and Deployment of an AI-based Student Slides Analysis tool for Academic Presentations	AISSA：基于AI的学生幻灯片分析工具，用于提升学术报告质量	large language model
11	CodeEvolve: LLM-Driven Evolutionary Optimization with Runtime-Enriched Target Selection for Multi-Language Code Enhancement	CodeEvolve：基于LLM和运行时信息的演化优化框架，用于多语言代码增强	large language model
12	VocalParse: Towards Unified and Scalable Singing Voice Transcription with Large Audio Language Models	VocalParse：基于大型音频语言模型的统一可扩展歌声转录方法	chain-of-thought	✅
13	SensingAgents: A Multi-Agent Collaborative Framework for Robust IMU Activity Recognition	提出SensingAgents多智能体框架，提升IMU活动识别的鲁棒性和零样本性能。	large language model
14	SADE: Symptom-Aware Diagnostic Escalation for LLM-Based Network Troubleshooting	SADE：基于LLM网络故障排除的症状感知诊断升级方法	large language model
15	A Hybrid Method for Low-Resource Named Entity Recognition	提出一种混合神经符号方法，解决低资源越南语命名实体识别问题。	large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#	题目	一句话要点	标签	🔗	⭐
16	Executable World Models for ARC-AGI-3 in the Era of Coding Agents	提出基于可执行世界模型的ARC-AGI-3编码智能体系统	world model world models
17	Pen-Strategist: A Reasoning Framework for Penetration Testing Strategy Formation and Analysis	Pen-Strategist：一种用于渗透测试策略生成与分析的推理框架	reinforcement learning penetration
18	Strat-Reasoner: Reinforcing Strategic Reasoning of LLMs in Multi-Agent Games	Strat-Reasoner：强化学习驱动LLM在多智能体博弈中的策略推理	reinforcement learning large language model chain-of-thought
19	Joint Optimization of Trajectory Control, Resource Allocation, and Task Offloading for Multi-UAV-Assisted IoV	针对多无人机辅助的车联网，提出联合优化轨迹控制、资源分配和任务卸载方案。	reinforcement learning deep reinforcement learning DRL
20	Reward-Decomposed Reinforcement Learning for Immersive Video Role-Playing	提出EBM-RL框架，解决沉浸式视频角色扮演中氛围和情感表达不足的问题	reinforcement learning

⬅️ 返回 cs.AI 首页 · 🏠 返回主页