cs.AI（2026-09-09）

📊 共 22 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (13 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (6) 支柱一：机器人控制 (Robot Control) (2) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (13 篇)

#	题目	一句话要点	标签	🔗	⭐
1	Beyond Surface Imitation: Contrastive Modeling for Reasoning Path Alignment in Multimodal In-Context Learning	提出对比建模框架以解决多模态学习中的推理路径对齐问题	large language model multimodal
2	OntologyAligner: Ontology-Aligned Retrieval and Hierarchy-Guided Large Language Model Reranking for Biomedical Ontology Normalization	提出OntologyAligner以解决生物医学本体规范化问题	large language model	✅
3	Structural Process Supervision for Latent Chain-of-Thought Reasoning	提出原型介导过程监督以解决潜在推理中的信息分布不均问题	chain-of-thought
4	Arbitrary Cipher Attacks Against Large Language Models Do Not Require Fine-Tuning	提出新攻击方法以破解大型语言模型的安全性	large language model
5	Belief-State Engine: Augmenting LLMs for Principled Planning Under Partial Observability	提出信念状态引擎以解决部分可观测环境下的规划问题	large language model chain-of-thought
6	From Symbolic Perception to Logical Deduction: A Framework for Guiding Language Models in Geometric Reasoning	提出一种框架以指导语言模型进行几何推理	large language model multimodal
7	RAP: Research Attention Prediction Reveals Target-Conditioned Evidence Acquisition Biases	提出研究注意力预测模型以解决研究关注度评估难题	large language model
8	Grounded Evaluation and Repair for NL-to-PDDL Problem Generation	提出基于LLM的NL到PDDL问题生成与修复方法	large language model
9	LexAgentHallu: A Hierarchical Benchmark for Profiling Hallucinations in Legal Agents	提出LexAgentHallu以解决法律代理人幻觉评估问题	large language model
10	PRAGMA: Evaluating Personalized Guidance with Memory Alignment in Lifelong Conversations	提出PRAGMA基准以解决个性化指导中的记忆对齐问题	large language model
11	From State Synchronization to Cognitive Self-Evolution: An Operational Architecture for Cognitive Digital Twins	提出认知数字双胞胎架构以解决现有数字双胞胎认知集成不足问题	large language model
12	Multi-Agent Agentic Graph Learning via Structural Signatures	提出MAAGL框架以解决图学习中的多代理协作问题	large language model
13	The Vibe Shift in Software Engineering: Evaluating AI-Led Conversational Programming for Performance, Cognition, and Responsible Adoption	提出Vibe编码以提升软件开发效率与责任采用	large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)

#	题目	一句话要点	标签	🔗	⭐
14	Decision Transformer for UAV-Mounted RIS-Assisted Dynamic D2D Communications	提出决策变换器以优化无人机辅助的动态D2D通信	reinforcement learning deep reinforcement learning DRL
15	FlowCPO: A Unified Divergence View of Preference Alignment for Flow Models	提出FlowCPO以统一流模型的偏好对齐问题	reinforcement learning flow matching DPO
16	Proof-Carrying Cognition: Closing the Verification Gap with Reality-Settled Reward	提出基于现实奖励的证明携带认知以缩小验证差距	reinforcement learning world model world models
17	ConvMem: Convolutional Memory for Long-Context Reasoning	提出ConvMem以解决长上下文推理问题	reinforcement learning large language model
18	Decision Shifts, Lost Label Functionality, and an Inconclusive Grounding Audit in Correctness-Gated Multi-Teacher Distillation	提出正确性门控多教师蒸馏以解决决策偏移问题	distillation
19	Seven Sources of Physical AI Capability Formation	提出七种来源以解析物理人工智能能力形成问题	world model world models predictive model

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#	题目	一句话要点	标签	🔗	⭐
20	Time-Frequency Geometric Cross-Attention for Chunked Vision-Language-Action Models	提出时间-频率几何交叉注意力以解决视觉-语言-动作模型的不足	domain randomization vision-language-action VLA
21	Modality-Decoupled Federated Learning for Privacy-Preserving Embodied Intelligence in 6G	提出FedMVLA框架以解决6G环境下隐私保护问题	manipulation vision-language-action VLA

🔬 支柱八：物理动画 (Physics-based Animation) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
22	Agent-Based ML-LLM Fusion with Self-Optimizing Prompts for Plateau Weather Alerts	提出SmartWeatherAgent以解决旅游气象服务中的适应性不足问题	spatiotemporal large language model

⬅️ 返回 cs.AI 首页 · 🏠 返回主页