cs.AI（2026-06-25）

📊 共 34 篇论文 | 🔗 5 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (18 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (11 🔗2) 支柱一：机器人控制 (Robot Control) (3 🔗2) 支柱五：交互与反应 (Interaction & Reaction) (1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (18 篇)

#题目一句话要点标签🔗⭐
1 Joint Learning of Experiential Rules and Policies for Large Language Model Agents 提出JERP以解决LLM代理在多步交互环境中的经验利用问题 large language model
2 A Pipeline for Generating Longitudinal Synthetic Clinical Notes Using Large Language Models 提出合成临床笔记生成管道以解决数据隐私问题 large language model
3 Content-Based Smart E-Mail Dispatcher Using Large Language Models 提出基于内容的智能邮件调度系统以解决邮件处理效率问题 large language model
4 TOPS: First-Principles Visual Token Pruning via Constructing Token Optimal Preservation Sets for Efficient MLLM Inference 提出TOPS以解决多模态大语言模型的视觉令牌冗余问题 large language model multimodal
5 TAVR-VLM: Risk-Conditioned Causal Grounding for Hallucination-Resistant Report Generation 提出TAVR-VLM以解决TAVR规划中的诊断幻觉问题 large language model multimodal
6 Scientific discovery as meta-optimization: a combinatorial optimization case study 提出元优化方法以提升科学发现效率 large language model
7 Language-Based Digital Twins for Elderly Cognitive Assistance 提出基于语言的数字双胞胎以解决老年人认知健康监测问题 large language model
8 Application of LLMs to Threat Assessment of Foreign Peacekeeping Missions 提出大型语言模型以解决外部维和任务的威胁评估问题 large language model
9 Where Do CoT Training Gains Land in LLM based Agents? 提出CoT训练方法以提升语言模型代理的行动预测能力 chain-of-thought
10 LCAi: Life Cycle Assessment with big data fusion and retrieval-augmented generation-assisted interpretation 提出基于大数据融合的LCA解读框架以应对环境热点问题 large language model
11 The Capability Frontier: Benchmarks Miss 82% of Model Performance 提出能力前沿以解决LLM评估不足问题 instruction following
12 EGG: An Expert-Guided Agent Framework for Kernel Generation 提出EGG框架以解决GPU内核生成中的优化指导问题 large language model
13 Do Safety Guardrails Need to Reason? LeanGuard: A Fast and Light Approach for Robust Moderation 提出LeanGuard以解决安全护栏推理效率低下问题 chain-of-thought ✅
14 Agents That Know Too Much: A Data-Centric Survey of Privacy in LLM Agents 提出数据中心化视角以解决LLM代理隐私问题 large language model
15 LLM-based Models for Detecting Emerging Topics in Service Feedback 提出基于LLM的模型以检测服务反馈中的新兴主题 large language model
16 An Empirical Study of LLM-Generated Specifications for VeriFast 评估LLM生成的规范以优化VeriFast的静态验证 large language model
17 Odyssey: Constructing Verifiable Local Truth-Preserving Foundation Models 提出ODYSSEY框架以构建可验证的本地真理保持基础模型 foundation model
18 A Pipeline for Generating Longitudinal Synthetic Clinical Notes Using Large Language Models 提出合成临床笔记生成管道以解决隐私问题 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🔗⭐
19 LithoDreamer: A Physics-Informed World Model for Multi-Stage Computational Lithography 提出LithoDreamer以解决多阶段计算光刻中的物理建模问题 world model world models dreamer ✅
20 EO-WM: A Physically Informed World Model for Probabilistic Earth Observation Forecasting 提出EO-WM以解决气象驱动的地球观测预测问题 world model world models ✅
21 Einstein World Models 提出爱因斯坦世界模型以增强大型语言模型的推理能力 world model world models
22 NebulaExp-8B: An Empirical Post-Training Pipeline via Full-Scale Ablation Research 提出NebulaExp-8B以解决大语言模型后训练透明性问题 reinforcement learning distillation large language model
23 A Latent ODE Approach to Spatiotemporal Modeling of Cine Cardiac MRI 提出潜在ODE模型以解决心脏MRI时空建模问题 latent dynamics spatiotemporal
24 Look-Before-Move: Narrative-Grounded World Visual Attention in Dynamic 3D Story Worlds 提出Look-Before-Move框架以解决动态3D环境中的视觉注意力问题 world model world models embodied AI
25 VoiceTTA: Enhancing Zero-Shot Text-to-Speech via Reinforcement Learning-Based Test-Time Adaptation 提出VoiceTTA以解决零-shot文本到语音合成中的风格模仿问题 reinforcement learning flow matching
26 EvoOptiGraph: Weakness-Driven Coevolution via Graph-Based Structural Generation for Optimization Modeling 提出EvoOptiGraph以解决优化建模中的数据与模型协同问题 reinforcement learning large language model
27 Internalizing the Future: A Unified Agentic Training Paradigm for World Model Planning 提出统一的代理训练范式以解决长远决策问题 reinforcement learning world model world models
28 Towards Evaluation of Implicit Software World Models in Coding LLMs 提出隐式软件世界模型评估方法以提升编码LLMs性能 world model world models
29 Look-Before-Move: Narrative-Grounded World Visual Attention in Dynamic 3D Story Worlds 提出Look-Before-Move框架以解决动态3D环境中的视觉注意力问题 world model world models embodied AI

🔬 支柱一：机器人控制 (Robot Control) (3 篇)

#题目一句话要点标签🔗⭐
30 Prompt Injection in Automated Résumé Screening with Large Language Models: Single and Multi-Injection Settings 研究提示注入以优化自动简历筛选中的候选人排名 manipulation large language model ✅
31 HiLSVA: Design and Evaluation of a Human-in-the-Loop Agentic System for Scientific Visualization 提出HiLSVA以解决科学可视化中的人机协作问题 manipulation large language model ✅
32 When Does Personality Composition Matter for Multi-Agent LLM Teams? 探讨个性组成对多智能体LLM团队表现的影响 manipulation large language model

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

#题目一句话要点标签🔗⭐
33 TGHE: Template-based Graph Homomorphic Encryption for Privacy-Preserving GNN Inference in Edge-Cloud Systems 提出TGHE以解决大规模金融图的隐私保护GNN推理问题 OMOMO

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

#题目一句话要点标签🔗⭐
34 XMSE-Aware Adaptive Empirical Bayes Estimation 提出XMSE感知自适应经验贝叶斯估计以优化参数估计 PULSE

⬅️ 返回 cs.AI 首页 · 🏠 返回主页