cs.AI（2026-08-21）

📊 共 33 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) (24 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (5) 支柱一：机器人控制 (Robot Control) (3) 支柱七：动作重定向 (Motion Retargeting) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (24 篇)

#题目一句话要点标签🔗⭐
1 ReFrame: Evidence-Guided Test-Time Safety Alignment in Multimodal Large Language Models 提出ReFrame以解决多模态大语言模型的安全对齐问题 large language model multimodal
2 Is Multimodal Speculative Decoding Ready for Diffusion-Based Parallel Drafting? A Survey and Empirical Diagnosis 提出多模态推测解码以解决扩散基础并行草拟问题 vision-language-action VLA multimodal
3 CellPath-Bench: A Multidimensional Benchmark for Whole-Slide Cellular Representations in Pathology Foundation Models 提出CellPath-Bench以评估病理基础模型的细胞表示能力 foundation model
4 Large Language Models at the Intersection of Software Engineering and Software Security:An Evidence-Centered Structured Survey and Research Agenda 提出证据中心结构调查以解决软件工程与安全交集问题 large language model
5 Evaluating Large Language Model Performance on International Maritime Dangerous Goods Code Compliance 提出DGEval基准以评估大型语言模型在IMDG合规性中的表现 large language model
6 Can Scientific Claims Be Removed from Large Language Models? A Systematic Evaluation of Claim-Level Unlearning 提出科学声明遗忘任务以解决语言模型知识过时问题 large language model
7 Foundation Models for Partial Causal Identification 提出因果基础模型以解决部分因果识别问题 foundation model
8 TLive-Omni: An Omni-Modal Understanding Model for E-Commerce Live Streaming 提出TLive-Omni以解决电商直播中的多模态理解问题 instruction following visual grounding
9 TRACE: Agentic Catalog Enrichment with Multi-source Evidence Grounding 提出TRACE框架以解决电商目录属性稀缺问题 large language model multimodal
10 When Generated Images Look Right and Retrieve Wrong: Coverage-Guided Cross-Scale Re-Indexing for Knowledge-Faithful Generative Perception 提出CERES框架以解决生成图像检索不一致问题 multimodal
11 Neuro-Geospatial Modelling of EEG Affective States Using Literature-Informed Environmental Context 提出基于文献信息的环境上下文的EEG情感状态建模方法 multimodal ✅
12 Trustworthy RAG: An Evaluation Agent for Detecting Misinformation and Knowledge Poisoning in Generative AI Systems 提出可信RAG以解决生成AI系统中的虚假信息和知识污染问题 large language model ✅
13 Can Legal AI Know When It Is Wrong? And Do Students Know When It Is? 提出高置信度错误率以解决法律AI过度自信问题 large language model
14 Belief Without Behavior: Measuring the Translation of Theory of Mind into Coordinated Social Action in Vision-Language Models 提出MOSAIC基准以解决理论心智与社会行为之间的脱节问题 multimodal
15 Deep Learning Models Also Recall Features 提出特征回忆概念以深化深度学习模型理解 large language model
16 Structured but Fragile: On the Limits of LLMs in Cybersecurity Decision-Making 探讨大型语言模型在网络安全决策中的局限性 large language model
17 Vibe Coding and Web Application Security: A Twin-Prompt Study 研究安全意识提示对生成Web应用程序安全性的影响 large language model
18 The Logic of Machine Self-Preservation 探讨机器自我保护逻辑以应对智能体行为问题 large language model
19 BC-Bench: Evaluating Agentic Engineering in a Domain-Specific Language for ERP 提出BC-Bench以评估ERP领域特定语言中的代理工程 multimodal
20 Certified Multi-Turn Robustness for LLM Safety via Compositional Bounds and Safety Persistence 提出多轮认证鲁棒性框架以提升LLM安全性 large language model
21 Profiling What Matters: Context-Aware Item Profiles from Large-Scale Metadata for LLM Recommenders 提出CAIRO框架以解决LLM推荐中的项目侧信息利用问题 large language model
22 DirEAG: Dirichlet Evidence Aggregation for Calibrating Verbalized Confidence in Mathematical Reasoning 提出DirEAG以解决数学推理中信心估计不可靠问题 large language model
23 VortexChat: An agentic framework for autonomous multi-objective integrated photonic design 提出VortexChat框架以实现自主的光子器件设计 large language model
24 Weighted Memory Tree: Remembering What Matters for Long-Horizon LLM Agents 提出加权记忆树以解决长时间跨度LLM代理的记忆问题 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🔗⭐
25 Advantage-level Aggregation Reinforcement Learning for X-point Target Magnetic Configuration Control in an EXL-50U Experiment-Calibrated Simulation Environment 提出Advantage级聚合强化学习以解决X点目标磁配置控制问题 reinforcement learning PPO
26 CDRL: Certification-Driven Reinforcement Learning for Neutrino Flavor Model Discovery 提出认证驱动强化学习以解决中微子味模型发现问题 reinforcement learning
27 $Z^2$-ACT: End-to-End Verifiable Agentic Intent Control for Open 6G RAN 提出$Z^2$-ACT以解决开放6G RAN中的多供应商控制问题 reinforcement learning large language model
28 Knowing but Not Saying: Preventing Factual Access Failures in LLM SFT via Recall-Anchored Distillation 提出Recall-Anchored Distillation以解决大语言模型的事实访问失败问题 distillation
29 Why2Speak: Faithful Reasoning for Abstaining Action Policies 提出Why2Speak以解决多方对话中的行动与沉默选择问题 reinforcement learning chain-of-thought

🔬 支柱一：机器人控制 (Robot Control) (3 篇)

#题目一句话要点标签🔗⭐
30 ForeTime-VLA: Causal Future-Token Distillation from a World Action Model for Conveyor-Belt Manipulation 提出ForeTime-VLA以解决动态物体操控中的未来预测问题 manipulation flow matching world model
31 Graph-Operator World Models for Morphology-Parameter Generalization in Continuous Control 提出Graph-Operator世界模型以解决形态参数泛化问题 MPC world model world models
32 Dynamic Context Scheduling: Learning Beyond the Static Universe 提出动态上下文调度以提升强化学习的泛化能力 bipedal biped

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

#题目一句话要点标签🔗⭐
33 SPARC: Single-Pass Scaling for Motion Forecasting with Conformal Bayesian Last Layers 提出SPARC以解决运动预测中的不确定性估计问题 human motion multimodal

⬅️ 返回 cs.AI 首页 · 🏠 返回主页