cs.AI（2026-06-04）

📊 共 53 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) (34 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (12 🔗1) 支柱一：机器人控制 (Robot Control) (4) 支柱五：交互与反应 (Interaction & Reaction) (1) 支柱三：空间感知与语义 (Perception & Semantics) (1) 支柱八：物理动画 (Physics-based Animation) (1 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (34 篇)

#题目一句话要点标签🔗⭐
1 TRACE: A Temporal Conditional Estimation for Multimodal Time Series Foundation Models 提出TRACE以解决多模态时间序列中的缺失与不对齐问题 foundation model multimodal
2 An Infectious Disease Spread Simulation Based on Large Language Model Decision Making 基于大语言模型决策的传染病传播模拟框架 large language model
3 Where Should Knowledge Enter? A Layered Framework for Knowledge Infusion in Multimodal Iterative Generative Mo 提出分层框架以解决多模态生成模型中的知识注入问题 multimodal
4 Amortizing Federated Adaptation: Hypernetwork Driven LoRA for Personalized Foundation Models 提出HyperLoRA以解决联邦学习中的适应性和聚合偏差问题 foundation model
5 Step-adaptive multimodal fusion network with multi-scale cloud feature learning for ultra-short-term solar irradiance forecasting 提出多源数据融合模型以解决超短期太阳辐射预测问题 multimodal
6 MLEvolve: A Self-Evolving Framework for Automated Machine Learning Algorithm Discovery 提出MLEvolve以解决机器学习算法发现中的信息孤岛问题 large language model ✅
7 Benchmark Everything Everywhere All at Once 提出Benchmark Agent以解决基准测试构建的可持续性问题 multimodal
8 Vortex: Efficient and Programmable Sparse Attention Serving for AI Agents 提出Vortex以解决稀疏注意力算法部署效率问题 large language model
9 TokenMizer: Graph-Structured Session Memory for Long-Horizon LLM Context Management 提出TokenMizer以解决长时间任务中的上下文管理问题 large language model
10 LLM Self-Recognition: Steering and Retrieving Activation Signatures 提出自我识别机制以增强大型语言模型的输出归属能力 large language model
11 ToolChoiceConfusion: Causal Minimal Tool Filtering for Reliable LLM Agents 提出Causal Minimal Tool Filtering以解决工具选择混淆问题 large language model
12 RedKnot: Efficient Long-Context LLM Serving with Head-Aware KV Reuse and SegPagedAttention 提出RedKnot以解决长上下文LLM服务中的KV缓存瓶颈问题 large language model
13 Towards the Readability of LLM-Generated Codes through Multitask Representation Engineering 提出多任务表示工程以提升LLM生成代码的可读性 large language model
14 Evaluating Agentic Configuration Repair for Computer Networks 提出基于代理配置修复的网络配置自动化方法 large language model
15 LLMCodec: Adapting Video Codecs for Efficient Weight Compression of Large Language Models 提出LLMCodec以解决大语言模型压缩问题 large language model
16 TAPO: Tool-Aware Policy Optimization via Credit Transfer for Multimodal Search Agents 提出TAPO以解决多模态搜索代理中的信用误分配问题 multimodal
17 HybridCodec: Fast Dual-Stream, Semantically Enhanced Neural Audio Codec 提出HybridCodec以解决音频编码中的语义信息引入问题 large language model multimodal
18 PerceptUI: LLM Agents as Human-Aligned Synthetic Users for UI/UX Evaluation 提出PerceptUI框架以提升UI/UX评估的效率与准确性 large language model multimodal
19 Seeing Time: Benchmarking Chronological Reasoning and Shortcut Biases in Vision-Language Models 提出新基准以评估视觉语言模型的时间推理能力 multimodal ✅
20 Critic-Guided Heterogeneous Multi-Agent Reasoning for Reliable Mathematical Problem Solving 提出基于评论指导的异构多智能体方法以提高数学问题求解的可靠性 large language model
21 Self-Commitment Latency: A Reward-Free Probe for Prompted Implicit Hacking 提出自承诺延迟以解决隐式奖励黑客问题 chain-of-thought
22 Mind the Gap: Bridging Behavioral Silos with LLMs in Multi-Vertical Recommendations 提出基于LLM的框架以解决多垂直推荐中的冷启动问题 large language model
23 MalTree: Tracing Malware Evolution from Embeddings at Scale 提出MalTree框架以自动化追踪恶意软件进化 TAMP
24 AI-Driven Test Case Generation from Natural Language Requirements: A Survey of Techniques and Research Gaps 提出基于AI的测试用例生成方法以解决自然语言需求的挑战 large language model
25 QCFuse: Query-Aware Cache Fusion via Compressed View for Efficient RAG Serving 提出QCFuse以解决RAG缓存融合效率问题 large language model
26 GenTI: Benchmarking LLMs for Autonomous IDPS Rule Generation for Unseen Attacks 提出GenTI以解决自动化IDPS规则生成的挑战 chain-of-thought
27 Queen-Bee Agents: A BeeSpec-Centered Architecture for Governed Enterprise MCP Orchestration 提出Queen-Bee架构以解决企业多代理系统治理问题 large language model
28 Microskill Architecture: A Modular Skill-Driven Framework for AI-Native Code Generation 提出MicroSkill架构以解决AI原生代码生成中的上下文管理问题 large language model
29 Agent-Orchestrated Adaptive RAG: A Comparative Study on Structured and Multi-Hop Retrieval 提出Agent-Orchestrated Adaptive RAG以解决复杂查询的检索问题 large language model
30 Enhancing Software Engineering Through Closed-Loop Memory Optimization 提出闭环记忆优化框架以提升软件工程代理的性能 large language model ✅
31 Evaluation of LLMs for Mathematical Formalization in Lean 评估大型语言模型在Lean中的数学形式化能力 large language model
32 Multilingual Fine-Tuning via Localized Gradient Conflict Resolution 提出基于局部梯度冲突解决的多语言微调方法 large language model
33 The End of Software Engineering: How AI Agents Are Fundamentally Restructuring the Software Paradigm 提出智能代理工程以重构软件工程范式 large language model
34 GuardNet: Ensemble Strategies of Shallow Neural Networks for Robust Prompt Injection and Jailbreak Detection 提出GuardNet以解决大语言模型的Prompt Injection和Jailbreak攻击问题 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (12 篇)

#题目一句话要点标签🔗⭐
35 WorldFly: A World-Model-Based Vision-Language-Action Model for UAV Navigation 提出WorldFly以解决无人机导航中的视角转变问题 flow matching world model world models
36 LatentWave: JEPA Pretraining for Wireless Foundation Models 提出LatentWave以解决无线任务模型偏差问题 JEPA Joint-Embedding Predictive Architecture joint-embedding predictive architecture
37 Learning to replenish: A hybrid deep reinforcement learning for dynamic inventory management in the pharmaceutical supply chains 提出混合深度强化学习以解决制药供应链动态库存管理问题 reinforcement learning deep reinforcement learning DRL
38 PLAN-S: Bridging Planning with Latent Style Dynamics for Autonomous Driving World Models 提出PLAN-S以解决自主驾驶中风险与风格动态建模问题 world model world models
39 Learning Visual Spatial Planning from Symbolic State via Modality-Gap-Aware Self-Distillation 提出MGSD框架以解决视觉空间规划中的模态差距问题 distillation multimodal ✅
40 TLA-Prover: Verifiable TLA+ Specification Synthesis via Preference-Optimized Low-Rank Adaptation 提出TLA-Prover以优化TLA+规范合成问题 DPO direct preference optimization large language model
41 Edit-R2: Context-Aware Reinforcement Learning for Multi-Turn Image Editing 提出Edit-R2以解决多轮图像编辑中的上下文保持问题 reinforcement learning flow matching foundation model
42 Towards World Models in Biomedical Research 提出生物医学世界模型以推动AI驱动的发现 world model world models large language model
43 Beyond Output Matching: Preserving Internal Geometry in NVFP4 LLM Distillation 提出CKA-QAD以解决低比特量化模型内部几何保留问题 distillation large language model
44 A Pre-Registered Causal Partition of Self-Consistency Elicitation and Reward Design in RLVR 提出自一致性引导与奖励设计的因果分解以优化RLVR reinforcement learning reward design
45 Statistical Priors for Implicit Preferences: Decoupling Skill Selection as a Local Harness in Personal Agents 提出轻量级本地偏好选择机制以解决个人智能体的用户偏好学习问题 preference learning large language model
46 Safety Paradox: How Enhanced Safety Awareness Leaves LLMs Vulnerable to Posterior Attack 提出后验攻击以揭示大型语言模型的安全悖论 reinforcement learning large language model

🔬 支柱一：机器人控制 (Robot Control) (4 篇)

#题目一句话要点标签🔗⭐
47 CogManip: Benchmarking Manipulative Behavior in Multi-Turn Interactions with Large Language Model 提出CogManip以评估大型语言模型中的操控行为风险 manipulation large language model
48 EEGDancer: Dynamic Emotion Latent Space Masked Modeling with Reinforcement Learning for EEG Continuous Emotion Prediction 提出EEGDancer以解决EEG连续情感预测问题 trajectory optimization reinforcement learning SAC
49 AEGIS: A Backup Reflex for Physical AI 提出AEGIS以解决长时间机器人操作中的失败问题 manipulation
50 TinyML-Driven Cybersecurity for Autonomous Spacecraft: Latency-Accuracy Analysis for SPARTA RF and Cyber Threat Detection 提出TinyML驱动的网络安全方案以解决自主航天器的网络威胁检测问题 manipulation

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

#题目一句话要点标签🔗⭐
51 Goedel-Architect: Streamlining Formal Theorem Proving with Blueprint Generation and Refinement 提出Goedel-Architect以简化形式定理证明过程 IMoS

🔬 支柱三：空间感知与语义 (Perception & Semantics) (1 篇)

#题目一句话要点标签🔗⭐
52 From Reward-Hack Activations to Agentic Risk States: Context-Calibrated Mechanistic Monitoring in LLM Agents 提出上下文校准的机制监控以解决大语言模型代理的安全性问题 affordance

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

#题目一句话要点标签🔗⭐
53 SagnacAssisted Enhanced OTDR for Distributed Acoustic Sensing: A Standardized Benchmark and Engineering Evaluation Framework 提出Sagnac辅助增强型φ-OTDR以解决分布式声学传感中的信号衰减问题 spatiotemporal ✅

⬅️ 返回 cs.AI 首页 · 🏠 返回主页