cs.AI（2026-09-03）

📊 共 29 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) (17 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (11 🔗1) 支柱四：生成式动作 (Generative Motion) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (17 篇)

#题目一句话要点标签🔗⭐
1 InSituMeasure: Probing Situated Measurement Grounding in Industrial Scenes with Multimodal Large Language Models 提出InSituMeasure以解决工业场景中的测量基础问题 large language model multimodal
2 NeoRed: A Knowledge-Logic-Alignment Multimodal Large Language Model for Neonatal Respiratory Disease Diagnosis 提出NeoRed以解决新生儿呼吸疾病诊断中的多模态数据整合问题 large language model multimodal
3 LLM4CKD: Large Language Models for Early Stage Chronic Kidney Disease Screening 提出LLM4CKD以解决慢性肾病早期筛查问题 large language model foundation model
4 IRWOZ 2.0: A Large Language Model-driven Dialogue Dataset for Industrial Robot Conversations 提出IRWOZ 2.0以解决工业人机对话系统的状态跟踪问题 large language model
5 Xiaomi-TabLDM: A Tabular Foundation Model Technical Report 提出Xiaomi-TabLDM以提升表格数据预测性能 foundation model
6 CulturalMenuBench: Probing the Knowledge-Application Gap in Multimodal Culinary Reasoning 提出CulturalMenuBench以解决多模态烹饪推理中的知识应用差距问题 multimodal
7 Do GUI Agents Know When Not to Act? Enabling Conflict-Aware Termination for Multimodal GUI Agents 提出CONFLICTGUARD以解决多模态GUI代理的冲突感知终止问题 multimodal
8 DuplexSpeechBench-IFEval: Evaluating Implicit Instruction Following in Full-Duplex Voice Agents 提出DuplexSpeechBench-IFEval以评估全双工语音代理的隐式指令遵循能力 instruction following
9 Epistemic Warrant for LLM Recommendations: Characterizing the Basis for Reliance When Ground Truth Is Unavailable 提出认知担保框架以解决LLM推荐信任问题 large language model
10 STAIR (STructure Aware Information Retriever): A novel dataset and LLM based retriever for document structure augmentation 提出STAIR以解决文档结构信息检索问题 large language model
11 SimSkill: A Lifelong Learning AI Agent for Autonomous Mastery of Traffic Simulation 提出SimSkill以实现交通仿真中的自主学习与能力提升 large language model ✅
12 Proactive Service Agents: A Unified Decision Framework, Methods, and Evaluation 提出主动服务代理以解决用户指令依赖问题 large language model
13 Can LLMs Extract Architectural Design Decisions from Source Code Commits? - A Preliminary Exploratory Study 利用大型语言模型提取源代码提交中的架构设计决策 large language model
14 HalluPeer: A Taxonomy-driven Benchmark for Detecting Hallucinations in Scientific Peer Reviews 提出HalluPeer以解决科学同行评审中的幻觉检测问题 large language model ✅
15 Dalek: A Constructive Agent Machine 提出Dalek以实现自我维护与自我进化的智能体机器 large language model
16 Caught in the Story: Narrative Captivity in Multi-turn LLMs Conversation 提出叙事囚禁概念以解决多轮对话中的道德判断偏差问题 large language model
17 A Prompt-Engineering Approach to Develop Scalable, Flexible, and Real-Time Hybrid Micro-Level Personalization in a General Purpose AI Teaching Assistant 提出基于提示工程的框架以实现个性化AI教学助手 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (11 篇)

#题目一句话要点标签🔗⭐
18 Rethinking On-Policy Distillation of Large Language Models II: One Training Example 提出单一查询的在线蒸馏方法以提升大语言模型性能 distillation large language model
19 Semantic Bayesian World Models 提出语义贝叶斯世界模型以解决知识图谱与语言模型的整合问题 world model world models foundation model
20 Rethinking World Models for Safety-Critical Embodied Systems 提出风险知情世界模型以解决安全关键系统决策问题 world model world models latent dynamics
21 From Prior-Guided Heuristics to Deployable Agents: Accelerating Demonstration-Driven Reinforcement Learning for Deadline-Constrained Network Control 提出基于有效拥塞的多智能体深度强化学习框架以解决网络控制中的时延问题 reinforcement learning deep reinforcement learning DRL
22 StrixAE: An Intelligent Agent for Audio Enhancement under Complex Distortion Coupling in Real-World Scenarios 提出StrixAE以解决复杂失真耦合下的音频增强问题 reinforcement learning reward design large language model
23 PPO-STGNN: A Proximal Policy Optimization Approach with Spatio-Temporal Graph Neural Networks for DAG Task Scheduling in Cloud-Edge-End Computing 提出PPO-STGNN以解决云边端计算中的DAG任务调度问题 reinforcement learning PPO behavior cloning
24 SENTINEL-RL: Offloading Topological Reasoning from LLM Agents in the Security Operations Center 提出SENTINEL-RL以解决大型语言模型在安全运营中心的局限性问题 PPO large language model
25 When Models Edit Too Much: On the Fidelity of Minimal Code Edits 提出最小化代码编辑方法以提高代码修复的准确性 reinforcement learning large language model
26 Synthetic Semantic Supervision for Contrastive Code Representation Learning in Small Transformers: An Empirical Study 提出合成语义监督以提升小型变换器的代码表示学习 representation learning
27 Symmetries and Causality: Causal Effect Identification Beyond IID Data 提出基于对称性的新方法以解决因果效应识别问题 reinforcement learning world model world models
28 Why Gated DeltaNet Survives 4-Bit Quantization: NVFP4 W4A4 for the Recurrent Half of a Hybrid 27B LLM 提出NVFP4 W4A4以解决混合27B LLM的4位量化问题 linear attention PULSE ✅

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
29 FLY-EVAL++: An Evidence-Driven Evaluation Protocol for Safety-Constrained Flight Prediction with Large Language Models 提出FLY-EVAL++以解决安全约束下的飞行预测评估问题 physically plausible large language model

⬅️ 返回 cs.AI 首页 · 🏠 返回主页