cs.AI（2026-10-07）

📊 共 31 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (20 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (7 🔗1) 支柱八：物理动画 (Physics-based Animation) (2) 支柱一：机器人控制 (Robot Control) (1) 支柱五：交互与反应 (Interaction & Reaction) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (20 篇)

#题目一句话要点标签🔗⭐
1 Reasoning-Token Spikes Under Prompted Untruthful Responding in Large Language Models 提出基于推理令牌的信号以检测大型语言模型的不真实响应 large language model chain-of-thought
2 Successive Training Stages and Large Language Model Persuasion: Effects of Misalignment, Supervised Fine-Tuning, and Preference Optimization 研究三阶段训练对大语言模型说服力的影响 large language model
3 DUDA-Bench: Benchmarking LLM Agents on Multimodal Data-Driven Urban Diagnosis 提出DUDA-Bench以解决城市诊断中的数据整合问题 multimodal
4 TopoGraphRAG-Bench: Evaluating Multimodal GraphRAG on Layout-Grounded Evidence Reasoning 提出TopoGraphRAG-Bench以解决多模态文档证据推理问题 multimodal ✅
5 Adversarial Images Hijack Web Agents from Visual Grounding to Browser Execution 提出WebMirage框架以解决视觉基础网页代理的控制问题 visual grounding
6 TaoD2C-Bench: Benchmarking MLLMs for Industrial UI Code Generation Beyond Visual Fidelity 提出TaoD2C-Bench以解决工业UI代码生成中的多模态推理问题 large language model multimodal
7 Open-MMUnlearning: Unifying Methods and Evaluation for MLLM Unlearning 提出Open-MMUnlearning以解决多模态大语言模型的去学习问题 large language model multimodal
8 RECAST: Learning to Compute the Right Context through Adaptive Evidence Routing 提出RECAST以解决多源信息证据构建问题 large language model
9 Validity Without Ground Truth: What Stated-Preference Economics Offers the Evaluation of Language Models 提出基于陈述偏好的经济学框架评估语言模型的有效性 large language model
10 SLDR: Defending Against Malicious Fine-tuning via Selective Layers Recovery and Dynamic Routing 提出SLDR以防御恶意微调问题 large language model ✅
11 LLM-Assisted Generation of Transparent, Open-Source Multiphysics Models of Electrochemical Devices 提出LLM辅助生成透明开源多物理场模型以解决电化学设备建模问题 large language model
12 Beyond LLM-GA: Secure Fluid Antenna Systems with ReEvo-Designed Memetic Algorithm 提出基于ReEvo设计的记忆算法以解决流体天线系统安全问题 large language model
13 Why Software Engineering Is Indispensable in the Age of Coding Agents 论证软件工程在智能编码时代的重要性 large language model
14 Agentic AI-Assisted Modeling for Production Scheduling: Assessment in Constraint Programming 提出通用LLM辅助约束编程模型以优化生产调度 large language model
15 UniSkill: Learning Actor-Aligned Skill Proposals for an Evolving Policy 提出UniSkill以解决技能提议与策略演化的协同问题 large language model ✅
16 From Expert-Guided Proof Search to Automated Open-Problem Solving 提出Bolzano系统以实现自动化开放问题求解 large language model
17 Cost-Efficient Theorem Proving via Agent Orchestration in Program Verification 提出CoCo-Prover以实现高效的程序验证 large language model
18 SafeEvo: Deciphering the Safety Alignment Mechanism and Evolution in Language Models 提出SafeEvo框架以解析语言模型的安全对齐机制 large language model
19 MARS: Malware Analysis with Rule-Based Scoring of LLM Claims 提出MARS框架以优化恶意软件分析的分类准确性 large language model
20 We Query, Therefore We Compute: On Oracle Computation beyond the Machine, with an Application to Agents 构建抽象机器以提升智能体系统的计算能力 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🔗⭐
21 RoboJEPA: Scaling Robotic Latent World Models 提出RoboJEPA以解决机器人潜在世界模型的扩展性问题 world model world models JEPA
22 SearchWorld: Spatial Value-Grounded Imagination for UAV Object Search via World Models 提出SearchWorld以解决无人机目标搜索中的部分可观测性问题 world model world models egocentric
23 TutorLoop: Regulating Student Learning Behaviors via Sensor-in-the-Loop Generative Feedback 提出TutorLoop以通过传感器反馈调节学生学习行为 reinforcement learning deep reinforcement learning DRL
24 A Scoping Review and Experimental Study on Reinforcement Learning from Human Feedback for Human-Robot Collaboration 提出基于人类反馈的强化学习方法以提升人机协作安全性 reinforcement learning RLHF
25 Learning to Accumulate Knowledge with Mutual Information 提出知识编织者以解决知识积累效率低下问题 reinforcement learning large language model ✅
26 Deadline-Aware Multi-Agent Reinforcement Learning for TSN-Based Vehicular Edge Networks 提出基于多智能体强化学习的调度方法以解决车载边缘网络中的延迟问题 reinforcement learning
27 MIMESIS: Learning User Simulators as Training Environments for Interactive Agents 提出MIMESIS以解决交互式代理训练中的用户模拟问题 reinforcement learning distillation

🔬 支柱八：物理动画 (Physics-based Animation) (2 篇)

#题目一句话要点标签🔗⭐
28 Learning Situation-Conditioned Thinking Policies for Long-Term LLM Agents 提出情境条件思维策略以解决长期LLM代理的记忆问题 spatiotemporal large language model
29 MeshSIPP: Efficient Lattice Planning in Dynamic Environment 提出MeshSIPP以解决动态环境中的高效路径规划问题 spatiotemporal

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
30 EmbodiedRSI: Active Continual Robot Learning Through Hypothesis-Guided Co-Evolution 提出EmbodiedRSI以解决机器人持续学习中的探索效率问题 teleoperation foundation model

🔬 支柱五：交互与反应 (Interaction & Reaction) (1 篇)

#题目一句话要点标签🔗⭐
31 Let the Library Speak: Self-Advertised Method Selection for Formal Proving 提出自我广告方法选择以解决形式证明中的适用性问题 IMoS

⬅️ 返回 cs.AI 首页 · 🏠 返回主页