cs.AI（2026-10-13）

📊 共 38 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (26 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (7) 支柱一：机器人控制 (Robot Control) (2) 支柱三：空间感知与语义 (Perception & Semantics) (2) 支柱七：动作重定向 (Motion Retargeting) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (26 篇)

#题目一句话要点标签🔗⭐
1 Neuro-Symbolic Graph Transformers for Hierarchical Intention Recognition from Multimodal Human Activity Sequences 提出神经符号图变换器以解决多模态人类活动序列的意图识别问题 multimodal
2 DTScribe: Generating Models for Digital Twin Reporting using Local Large Language Models 提出DTScribe以自动化数字双胞胎报告模型生成 large language model
3 Rethinking Quality in Large Language Model Routing: Security-Conditioned Model-and-Budget Selection 提出SCRouter以解决大语言模型路由中的安全性与效用问题 large language model
4 The Smart Buildings Challenge: Learning Foundation Models for Real-World Optimization at Scale 提出建筑基础模型以解决建筑优化难题 foundation model
5 FedGraph: Defending Federated Large Language Model Fine-Tuning Against Backdoor Attacks via Graph-Based Aggregation 提出FedGraph以解决联邦大语言模型微调中的后门攻击问题 large language model
6 How Reliable Are Large Language Models for Soil Organic Carbon Assessment? An Australian Benchmark 评估大型语言模型在土壤有机碳评估中的可靠性 large language model
7 Chain of Thought Monitoring Can Fail Even When CoT Is Necessary 提出四级层次结构以提升Chain of Thought监控的可靠性 chain-of-thought
8 Beyond Black Boxes: Visual Grounding for Text Embeddings 提出基于草图的文本嵌入可视化方法以提升信息检索能力 visual grounding
9 Read the Room Before Your Model Generates 提出Scenario-Actor框架以优化语言模型的温度选择 large language model instruction following
10 Mitigating Cross-Modal Hallucinations by Correcting Modality Misrouting in Omnimodal LLMs 提出CaRe方法以解决多模态大语言模型中的模态误导问题 large language model multimodal
11 LLM Agents for Automated Post-Training via Structured Recipe Optimization 提出结构化配方优化以实现大语言模型的自动后训练 large language model ✅
12 Training LLMs for Multi-Agent Collaboration with Verifiable Information Rewards 提出CoSIR以解决多智能体协作中的信息共享问题 large language model
13 Degrees of Freedom in Multi-Agent Coordination: Between Centralized and Decentralized Regimes 提出自由度框架以优化多智能体系统协调问题 large language model
14 SolBehave: Evaluating the Behavioural Correctness of LLM-Generated Smart Contracts 提出SolBehave以评估LLM生成智能合约的行为正确性 large language model
15 Configuring RAG Systems: What Breaks, Where Originates, and How to Fix 提出系统化配置分析以解决RAG系统配置问题 large language model
16 A Global View of How LLMs Value Human Time 探讨大型语言模型如何评估人类时间价值 large language model
17 How Well Do LLMs Perform in Peer Review? Evidence from Finance and Accounting 探讨大型语言模型在同行评审中的有效性 large language model
18 Choosing LLMs for Strategic Agent Interactions 提出战略选择框架以优化LLM在博弈中的表现 large language model
19 Efficient Safety Guardrails through Latent Dual-Hypothesis Reasoning with On-Demand Trace Reconstruction 提出LR-Guard以提高安全决策效率与可解释性 large language model
20 Output Discipline Sets the Floor: A Falsification-Guided, Model-Agnostic Agent Protocol for Auditable Hypothesis-Driven Decision Making 提出模型无关的代理协议以增强假设驱动的决策过程 large language model
21 Beyond Answer Sharing: TrueMAD for Targeted Peer Engagement in Multi-Agent Debate 提出TrueMAD以解决多代理辩论中的参与不足问题 large language model
22 Fresh at Decision, Stale at Action: Routing New Information to LLM Decisions Already Made 提出时间戳理论以优化LLM决策的准确性与时效性 TAMP
23 Can Agents Forget When Asked? A Benchmark for Memory Unlearning in LLM Agents 提出Bemula基准以解决大语言模型代理的记忆遗忘问题 large language model ✅
24 SpecCloak: Probing Harmful Code Generation through Engineering Specifications 提出SpecCloak框架以探测有害代码生成问题 large language model
25 ECMR: Evidence-Aware Deliberation Control for Multi-Agent LLM Reasoning 提出ECMR以解决多智能体LLM推理中的语义假共识问题 large language model
26 Diff Diffusion: Predicting Edit Paths with Diffusion Models for Efficient Code Editing 提出Diff Diffusion以解决代码编辑中的高效编辑路径预测问题 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (7 篇)

#题目一句话要点标签🔗⭐
27 Thinker-WM: A Unified World Model with Shared Multimodal Representations for Embodied Intelligence 提出Thinker-WM以解决多模态世界模型的碎片化问题 world model world models metric depth
28 DPOison-VL: Clean Preference Poisoning in Vision-Language Direct Preference Optimization 提出DPOison-VL以解决多模态偏好数据安全问题 DPO direct preference optimization multimodal
29 Order-Consistent Latent Geometry for Reliable Short-Horizon Planning in World Models 提出自接近性条件以解决短期规划中的潜在几何问题 world model world models
30 Optimizing Team Behavior via Extremum-seeking Control in Multi-agent Reinforcement Learning 提出自适应多样性控制以优化多智能体强化学习中的团队行为 reinforcement learning
31 PaceReasoner: Label-Free Learning to Answer Briefly from Long Reasoning 提出PaceReasoner以提高短时推理的准确性 reinforcement learning chain-of-thought
32 Beyond the Verdict: Agentic Code Judging with Verifiable Feedback 提出ARBITER以解决编程竞赛中的反馈不足问题 reinforcement learning large language model
33 Can an Agent Learn to Predict Its Own Failure? Outcome-Grounded Confidence Calibration in LLM Agents 提出基于结果的信心校准方法以提升LLM代理的决策能力 reinforcement learning offline RL

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
34 TacitVLA: Reasoning Vision-Language-Action Models Without Reasoning Latency 提出TacitVLA以解决视觉-语言-动作模型的推理延迟问题 manipulation vision-language-action VLA
35 Reference Wording Shapes Retrieval of Robot Task Instructions 提出基于参考描述的机器人任务指令检索方法 manipulation

🔬 支柱三：空间感知与语义 (Perception & Semantics) (2 篇)

#题目一句话要点标签🔗⭐
36 Back to the Future: Reverse-Consequence Modeling for Long-Horizon Particle Dynamics 提出ReStep以解决长时间粒子动力学预测问题 3D gaussian splatting gaussian splatting splatting
37 ABot-ER: Unified Embodied Reasoning through Agentic Data-Model Co-Evolution 提出ABot-ER以解决统一体智能推理的广度与深度问题 affordance foundation model

🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)

#题目一句话要点标签🔗⭐
38 GeoMER: A Joint Euclidean-Hyperbolic Geometry Framework for Multimodal Emotion Recognition 提出GeoMER框架以解决多模态情感识别中的层次结构问题 motion representation multimodal

⬅️ 返回 cs.AI 首页 · 🏠 返回主页