cs.AI（2026-08-07）

📊 共 40 篇论文 | 🔗 7 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (27 🔗6) 支柱二：RL算法与架构 (RL & Architecture) (11 🔗1) 支柱六：视频提取与匹配 (Video Extraction) (1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (27 篇)

#题目一句话要点标签🔗⭐
1 AgentPatch: Coarse-to-Fine Weak-Task Repair for Merging Agentic Multimodal Large Language Models 提出AgentPatch以解决多模态大语言模型合并中的弱任务修复问题 large language model multimodal ✅
2 Recipes for Creativity: Iterative Generation and Evaluation in Large Language Models 提出迭代生成与评估方法以提升大型语言模型的创意表现 large language model
3 Unsupervised Adaptation of PDE Foundation Models 提出无监督PDE基础模型适应方法以解决数据稀缺问题 foundation model
4 ZIPBrain: Can EEG Foundation Models Be Faster, Locally Deployable, but Accurate? 提出ZIPBrain以解决EEG模型速度与准确性问题 foundation model
5 Critical Acclaim Orientation in Large Language Models: Evidence from Film Preference Elicitation 探讨大型语言模型的电影评价偏向性问题 large language model
6 Gated-BEPO: Confidence-Gated Bellman Credit Assignment for Large Language Model Agents 提出Gated-BEPO以解决长时间环境中的信用分配问题 large language model
7 LoRAScan: Detecting Backdoor Prompts in Low-Rank Adapters for Large Language Models via Down-Projection Activation Spikes 提出LoRAScan以检测低秩适配器中的后门提示 large language model
8 Progressive Alignment of Recommender Foundation Model through Multi-Phase Post-Training 提出三阶段渐进式后训练框架以优化推荐系统 foundation model ✅
9 CellWorld: From Gene-Level Reconstruction to Latent Cell Prediction in Spatial Transcriptomics Foundation Models 提出CellWorld以解决空间转录组基础模型的重建问题 foundation model ✅
10 Toward a Causal Data Management Ecosystem for Decision Making and Agentic AI 提出因果数据管理生态系统以解决决策与自主智能问题 large language model multimodal
11 Science Edge Evaluation: SEE the Missing Step Toward Real Scientific Discovery 提出科学边缘评估方法以解决科学发现中的推理不足问题 large language model multimodal
12 GeoBenchLLM: A Comprehensive Benchmark for Evaluating LLMs on Geo-Related Tasks 提出GeoBenchLLM以评估大语言模型在地理任务上的表现 large language model ✅
13 Strategy-first synthesis planning for complex natural products 提出SynthEx框架以解决复杂天然产物合成规划问题 large language model
14 Blast Radius 提出Blast Radius以解决代理编码中的代币消耗问题 large language model
15 CoBa: Cost-Effective Test-Time Scaling via Compute-Balanced Routing 提出CoBa以解决测试时间计算资源分配问题 chain-of-thought
16 PACE: Primitive-Aware Code Evolution for Automated Algorithm Design 提出PACE以解决算法设计中的局部逻辑耦合问题 large language model
17 People Are Not Just Their Countries. Disentangling Social Determinants of LLM Value Alignment Across Europe 通过社会人口变量解析LLM价值对齐问题 large language model
18 An End-to-End Agent Auditing Engine 提出A^2E以解决代理评估体系构建挑战 large language model ✅
19 SCALE: Scientific Concept Aggregation via LLMs and Embeddings for Fine-Grained Taxonomy Extension 提出SCALE框架以解决科学研究分类系统的细粒度问题 large language model
20 SetEasy: A Multi-Modal Classroom Engagement Assessment and Seating Optimization Framework 提出SetEasy框架以优化课堂参与度和座位安排问题 multimodal
21 ReQuant: Fixed-Grid Discrete Refinement for Post-Training Quantization 提出ReQuant以优化后训练量化模型的权重分配 large language model
22 LMM Modality Transfer: A Pre-requisite for Autonomous GIS Agents 提出多模态转移方法以解决GIS代理的空间理解问题 multimodal
23 Long-Horizon Agent Trajectory Attribution: A Unified Benchmark and Fine-Grained Annotation Framework 提出轨迹归因基准与细粒度注释框架以解决长时间跨度代理分析问题 large language model ✅
24 Coupling Planning with Episodic Memory in LLM Agents for Software Issue Resolution 提出PMCoder以结合规划与记忆解决软件问题 large language model
25 Evolving Parallel Algorithm Portfolios via Potential-Aware Instance Generation with LLMs 提出潜力感知实例与算法共演框架以提升组合优化问题求解能力 large language model
26 HLSmith: An Expert-Guided Agentic Framework for C/C++-to-HLS Translation 提出HLSmith框架以解决C/C++到HLS转换的专家知识缺乏问题 large language model
27 Mind the Gap: A Dual Knowledge Graph Framework for Unified Multi-task User Intent Inference 提出DKG-MTI框架以解决多任务用户意图推断问题 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🔗⭐
28 Surg-UniWorld: A Unified Surgical World Model with Multimodal Control Experts 提出Surg-UniWorld以解决手术世界模型的多模态控制问题 world model world models optical flow
29 Transformers Struggle to Use Their Emergent World Models: Revisiting the Tower of Hanoi, and the Illusion of Thinking 提出新方法分析变压器在汉诺塔问题中的表现 world model world models chain-of-thought
30 A MARL Centered Reference Architecture for Large Language Model Augmentation in Smart Manufacturing 提出MARL中心参考架构以增强智能制造中的大语言模型 reinforcement learning reward design large language model
31 Vehicle routing problem using deep reinforcement learning - A case study about truck planning in the industry 基于深度强化学习的车辆路径优化解决物流调度问题 reinforcement learning deep reinforcement learning DRL
32 MemWM: Memory-Augmented Text-Based World Model 提出MemWM以解决文本基础世界模型的系统性预测错误问题 world model world models
33 ResidencyRL: Reinforcement Learning in Simulated Clinical Environments 提出ResidencyRL以优化临床决策过程中的强化学习 reinforcement learning large language model
34 IB-RL: Isolated Bilateral Reinforcement Learning for Strategic Dialogue Agents 提出IB-RL以解决静态对手不匹配问题 reinforcement learning large language model
35 How Much, Then Where: Credit-Conserving Action-to-Token Allocation for Multi-Turn Agent Reinforcement Learning 提出FACTOR以解决多回合强化学习中的信用分配问题 reinforcement learning teacher-student
36 MemOPD: On-Policy Distillation through Memory State Alignment for Long-Horizon Agents 提出MemOPD以解决长时间代理的记忆状态对齐问题 PPO distillation ✅
37 Agent Memory Distillation: Empowering Small LLM Agents with Hierarchical Teacher Memory 提出Agent Memory Distillation以提升小型LLM代理的性能 distillation
38 Capek 0.5: An Execution-Centric Vision-Language Model for Embodied Intelligence 提出Capek 0.5以解决机器人执行过程中的多模态理解问题 reinforcement learning distillation

🔬 支柱六：视频提取与匹配 (Video Extraction) (1 篇)

#题目一句话要点标签🔗⭐
39 WNM-3D: A World Navigation Model with 3D Scene Conditioning for Closed-Loop VLN 提出WNM-3D以解决闭环视觉语言导航中的场景建模问题 egocentric vision-language-action VLA

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#题目一句话要点标签🔗⭐
40 ReGraph: Learning to Generate Recipe Graphs from Food Images 提出ReGraph以解决食物图像生成食谱图的结构化问题 manipulation multimodal chain-of-thought

⬅️ 返回 cs.AI 首页 · 🏠 返回主页