cs.AI（2026-07-29）

📊 共 26 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) (18 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (5 🔗1) 支柱一：机器人控制 (Robot Control) (3)

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

#题目一句话要点标签🔗⭐
1 MultivationBench: A Benchmark for Multimodal Sequential Motivation Reasoning 提出MultivationBench以解决多模态序列动机推理评估问题 large language model multimodal
2 Human diversity fuels collective creativity that large language models cannot simulate or sustain 探讨人类多样性如何推动创造力，AI无法模拟 large language model
3 Scientific Knowledge Discovery in the Age of Large Language Models 利用生成性大语言模型提升学术文献检索与筛选效率 large language model
4 WhisperRec: Latent Reasoning for Efficient Foundation Recommendation Models 提出WhisperRec以解决推荐模型推理效率问题 large language model chain-of-thought
5 Crossing-Free Probabilistic K-Line Forecasts Without Retraining 提出K线-分位序列投影方法以解决K线预测中的一致性问题 foundation model
6 Rethinking Self-Evolution: A Constrained Exploration-Exploitation Process for Mitigating Skill Overfitting 提出SkillBoost以解决技能过拟合问题 large language model
7 LLMET: Enabling Cross-Layer Evaluation of Emerging M3D Memories for Energy-Efficient LLM Serving 提出LLMET框架以提升大语言模型的能效表现 large language model
8 EvoPINN: Agentic Discovery of Executable Algorithms for Physics-Informed Neural Networks 提出EvoPINN以解决物理信息神经网络算法发现问题 large language model
9 OmegaUse-OfficeVal: Benchmarking LLM Agents on Long-Horizon Office-Suite Tasks with Economic Grounding 提出OmegaUse-OfficeVal基准以评估LLM在长时间办公任务中的经济效益 large language model
10 AgentMap: Joint Equivalence and Subsumption Discovery for Ontology Matching 提出AgentMap以解决本体匹配中的等价与包含关系发现问题 large language model
11 MMAC: A Massive Multi-dimensional Benchmark for Audio Captioning 提出MMAC基准以提升音频描述的评估质量 large language model
12 SecRespond: Benchmarking AI Agents for Real-World Post-Compromise Incident Response 提出SecRespond以解决后渗透事件响应评估问题 large language model ✅
13 Do Latent Channels Actually Communicate? A Causal Audit of Latent Multi-Agent LLM 提出因果审计方法以评估潜在多智能体LLM的通信效果 large language model
14 UrbanDS: A Graph-Guided LLM Multi-Agent System for Data-Intensive Urban Tasks 提出UrbanDS以解决数据密集型城市任务的挑战 large language model
15 AgenticCANN: Automated Ascend C Operator Generation via Knowledge-Augmented Agentic Evolution 提出AgenticCANN以解决Ascend C操作符自动生成问题 large language model
16 Living-Harness Is an Interactive-Agent Evolver 提出Living-Harness以解决交互代理的执行失败问题 large language model
17 Eco3S: Complex Socio-Economic System Simulation via Agent-Based Models 提出Eco3S框架以解决复杂社会经济系统模拟问题 large language model
18 PUDA: An AI-Native Hardware Harness for Self-Driving Laboratories 提出PUDA以解决自驾实验室的硬件交互问题 TAMP

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
19 CG-World: A Large-Scale World-State Dataset and Protocol for World Models 提出CG-World数据集以解决世界模型学习中的结构性不足问题 world model world models spatiotemporal
20 CaM-Wolf: Causal-Aware Multimodal Agents for Social Deduction Games 提出CaM-Wolf以解决社交推理游戏中的多模态交互问题 reinforcement learning large language model multimodal ✅
21 On-Policy Distillation for LLM Safety: A Routing Approach to Template-Robust Realignment 提出基于路由的在线蒸馏方法以解决LLM安全性问题 distillation large language model
22 Improving Item Discoverability in e-Commerce Search via Related Intent Generation 提出基于意图生成的搜索系统以提升电商商品可发现性 teacher-student distillation large language model
23 Graph Is the Verifier: Agentic Reinforcement Learning for Interprocedural Vulnerability Detection 提出VulAgentRL以解决跨过程漏洞检测问题 reinforcement learning

🔬 支柱一：机器人控制 (Robot Control) (3 篇)

#题目一句话要点标签🔗⭐
24 From Passive Video to Editable Experience: Physically Grounded Experience Synthesis for Embodied Intelligence 提出Pegasus框架以解决机器人学习人类操作视频的数据鸿沟问题 manipulation affordance egocentric
25 Guarding Organizations Against Malware Risk: A Novel Graph-Based Malware Detection Method 提出MalGuard以解决组织恶意软件风险管理问题 manipulation representation learning
26 What Does It Take to Detect an AI Agent? Minimal Feature Sets for Behavioral Detection under Browser Automation 提出三类检测框架以区分人类、机器人和AI代理 manipulation

⬅️ 返回 cs.AI 首页 · 🏠 返回主页