cs.AI（2026-06-08）

📊 共 30 篇论文

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (20) 支柱二：RL算法与架构 (RL & Architecture) (8) 支柱六：视频提取与匹配 (Video Extraction) (1) 支柱三：空间感知与语义 (Perception & Semantics) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (20 篇)

#题目一句话要点标签🔗⭐
1 Late-Layer Fusion is Enough: Dual-Path Vision Token Routing for Multimodal Large Language Models under Visual Saturation 提出双路径视觉令牌路由以解决多模态大语言模型的视觉饱和问题 large language model multimodal
2 IMUG-Bench: Benchmarking Unified Multimodal Models on Interleaved Understanding and Generation 提出IMUG-Bench以解决多轮图文对话评估问题 multimodal chain-of-thought
3 FMplex: Model Virtualization for Serving Extensible Foundation Models 提出FMplex以解决模型服务中的资源浪费问题 foundation model multimodal
4 Optical Reasoning: Rethinking Images as an Expressive Reasoning Medium Beyond Text 提出光学推理以解决多模态推理效率问题 large language model multimodal chain-of-thought
5 Pretrained, Frozen, Still Leaking: Auditing Cross-Encoder Attribute Transfer in EEG Foundation Models 提出跨编码器属性转移审计框架以解决EEG模型安全性问题 foundation model
6 Unveiling Privacy Risks in Multi-modal Large Language Models: Task-specific Vulnerabilities and Mitigation Challenges 提出MM-Privacy数据集以解决多模态大语言模型隐私风险问题 large language model
7 RTL-BenchLS: A Large-Scale Benchmark for RTL Reasoning and Generation with Large Language Models 提出RTL-BenchLS以解决现有RTL基准的规模与任务局限问题 large language model
8 TABVERSE: Benchmarking Cross-Format Table Understanding in LLMs and VLMs 提出TABVERSE以解决表格理解中的表示问题 large language model multimodal
9 Proxy Reward Internalization and Mechanistic Exploitation: A Learned Precursor to Reward Hacking and Its Generalization 提出PRIME以解决代理奖励黑客问题 chain-of-thought
10 SearchSwarm: Towards Delegation Intelligence in Agentic LLMs for Long-Horizon Deep Research 提出SearchSwarm以解决长时域深度研究中的任务委派智能问题 large language model
11 (Auto)formalization is supposed to be easy: Trellis process semantics for spelling out rigorous proofs 提出Trellis系统以简化自动形式化证明过程 generalist agent
12 FuseFSS: Efficient Secure LLM Inference with Function Secret Sharing 提出FuseFSS以提升安全LLM推理效率 large language model
13 Context-Aware Deep Learning for Defect Classification in Atomic-Resolution STEM 提出上下文感知深度学习框架以解决缺陷分类问题 multimodal
14 MASS: Deep Research for Social Sciences with Memory-Augmented Social Simulation 提出记忆增强社会模拟以提升社会科学研究的创造力 large language model
15 Steganography Without Modification: Hidden Communication via LLM Seeds 提出无修改的隐写通信方法以利用LLM种子 large language model
16 ComplexConstraints and Beyond: Expert Rubrics for RLVR 提出专家评分标准以提升RLVR评估方法的有效性 instruction following
17 Graph2Idea:Retrieval-Augmented Scientific Idea Generation with Graph-Structured Contexts 提出Graph2Idea以解决科学研究创意生成中的文献关系识别问题 large language model
18 LATTEArena: An Evaluation Framework for LLM-powered Tabular Feature Engineering (Extended Version) 提出LATTEArena以解决LLM驱动的表格特征工程评估问题 large language model
19 The Token Not Taken: Sampling, State, and the Variability of AI Agent Outputs 提出分层分析以解决AI代理系统输出变异性问题 foundation model
20 An Effective Router for Vision-Language Model Selection 提出ARMS路由器以解决视觉语言模型选择问题 multimodal

🔬 支柱二：RL算法与架构 (RL & Architecture) (8 篇)

#题目一句话要点标签🔗⭐
21 FF-JEPA: Long-Horizon Planning in World Models with Latent Planners 提出FF-JEPA以解决长时间规划中的目标图像依赖问题 world model world models JEPA
22 A Regret Minimization Framework on Preference Learning in Large Language Models 提出基于遗憾最小化的偏好优化方法以提升语言模型训练效果 reinforcement learning preference learning RLHF
23 AlloSpatial: Agentic Harness Framework for Spatial Reasoning in Foundation Models 提出AlloSpatial框架以解决多模态基础模型的空间推理问题 reinforcement learning egocentric foundation model
24 AliyunConsoleAgent: Training Web Agents in Real-World Cloud Environments via Distillation and Reinforcement Learning 提出AliyunConsoleAgent以解决云环境中文档验证问题 reinforcement learning distillation instruction following
25 Diverse Thinking Schemata Elicit Better Reasoning in Large Language Models 提出多样化思维模式优化以提升大型语言模型推理能力 reinforcement learning large language model
26 Personalization Meets Safety:Mechanisms,Risks,and Mitigations in Personalized LLMs 提出安全意识的个性化LLM评估框架以解决安全风险问题 reinforcement learning large language model multimodal
27 Baichuan-M4: A Clinical-Grade Medical Agent System for Continuous Care 提出Baichuan-M4以解决连续医疗护理问题 reinforcement learning curriculum learning multimodal
28 Next-Token Prediction Learns Generalisable Representations of Sleep Physiology 提出Hypnos模型以解决多模态生理信号表示学习问题 representation learning foundation model

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

#题目一句话要点标签🔗⭐
29 SpatialWorld: Benchmarking Interactive Spatial Reasoning of Multimodal Agents in Real-World Tasks 提出SpatialWorld以解决多模态智能体的空间推理评估问题 egocentric large language model multimodal

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

#题目一句话要点标签🔗⭐
30 I Was Scrolling and Then I Saw a Pregnant Strawberry 探讨AI迷你剧中的性别与种族叙事结构 affordance

⬅️ 返回 cs.AI 首页 · 🏠 返回主页