cs.AI（2026-09-15）

📊 共 31 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) (21 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (8) 支柱三：空间感知与语义 (Perception & Semantics) (2)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (21 篇)

#题目一句话要点标签🔗⭐
1 Layers, Sinks, and Scaling: Adaptive Evidence Selection for Multimodal Large Language Models 提出自适应证据选择方法以提升多模态大语言模型性能 large language model multimodal
2 RepoAtlas: Guiding Coding Agents via Evolving Multimodal Repository Views 提出RepoAtlas以解决代码库问题定位与上下文维护难题 large language model multimodal
3 Do LLMs Have Values? A Quantitative Analysis and Alignment Framework for Values in Large Language Models 提出PEC框架以解决大语言模型价值观对齐问题 large language model chain-of-thought
4 Integrating the Analytic Hierarchy Process with Large Language Models for Transparent Multi-Criteria Decision-Making 将层次分析法与大语言模型结合以提升决策透明度 large language model
5 VideoMM: Adaptive Macro-Micro Inference for Efficient Video MLLMs 提出VideoMM以解决长视频理解中的视觉令牌爆炸问题 large language model multimodal ✅
6 Sparse MLLM Anchors, Dense Adaptation: Breaking the Self-Referential Loop in Wild Test-Time Adaptation 提出MASA方法以解决野外测试时适应中的自我参考循环问题 large language model multimodal
7 little m: An AI Agent for Industrial Process Optimization 提出little m以解决工业过程优化建模问题 large language model multimodal ✅
8 Bridging Learned Visual Perception and Symbolic Belief-Space Planning 提出VLM作为概率基础的规划方法以解决不确定性问题 foundation model
9 CoAdapt: An LLM-based Framework for Adaptive Collaborative Perception in IIoT Robotic Swarms 提出CoAdapt框架以解决IIoT机器人群体适应性协作感知问题 large language model
10 Neuro-Symbolic Hierarchical Intention Anticipation in Human Behavior 提出神经符号层次意图预测以解决人类行为理解问题 multimodal
11 LimiX-2: A Contextual Mechanism Network Towards General Structured-Data Intelligence 提出LimiX-2以解决结构化数据智能问题 foundation model
12 Self-Emergence Agent Architecture:Behavior-Inertia HMM, Reflexive Metacognition,and Social-Contrastive Self-Modeling 提出自涌现代理架构以解决LLM代理的结构性局限 large language model
13 From Transient Prompts to Persistent Control: Scientific Poster Generation via Recursive Semantic-Geometric Contracts 提出PosterVisor框架以解决科学海报生成中的控制问题 multimodal
14 End-to-End Latency-Minimizing and Load-Balanced Request Scheduling for Edge LLM Inference in Agentic AI Services 提出LYREO方法以解决边缘LLM推理中的请求调度问题 large language model
15 FirmCORe: A Benchmark for Structured Reasoning about Inter-Firm Collaboration Opportunities 提出FirmCORe以解决企业间合作机会识别问题 large language model
16 FlexEE: Self-Speculative and KV-Compatible Early Exiting for Offloading-Aware LLM Inference 提出FlexEE以解决资源受限下的LLM推理效率问题 large language model
17 Can We Do Interpretable NLI with Graphs Based on Atomic Propositions? 提出图基推理方法以实现可解释的自然语言推理 large language model
18 Available but Unclaimed: An Empirical Study of Human-AI Synergy 研究人机协同，提升大语言模型的应用效果 large language model
19 A Framework for Generating Valid Context-Specific Benchmarks through Expert Guidance 提出一种框架以生成有效的上下文特定基准数据集 large language model
20 Protocol-Preserving Context Trimming for Agentic Workflows: Benefits, Failure Regimes, and Budget Guardrails 提出协议保留的上下文修剪以提高多步骤代理工作流的可靠性 large language model
21 Fine-Tuning Fixes Mode Collapse and Over-Dispersion in LLMs 提出细化调优以解决大语言模型的模式崩溃与过度分散问题 large language model

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

#题目一句话要点标签🔗⭐
22 SKIP: a Self-knowledge-guided Step-wise Preference Learning Framework for Concise Reasoning 提出SKIP框架以解决大语言模型推理效率低下问题 preference learning DPO large language model
23 AI for Games in the Foundation Model Era 提出基础模型以重塑游戏AI的生命周期 world model world models foundation model
24 Extracting ontology-compliant knowledge from scientific text describing irradiated materials using large language models 提出eolas以解决科学文献中知识提取的挑战 predictive model large language model
25 MOCC-R1: Reinforcing Reasoning-Response Consistency for Multimodal Counselor Response Generation 提出MOCC-R1以解决多模态咨询响应生成中的一致性问题 reinforcement learning multimodal
26 Intrinsic Motivation in Reinforcement Learning: A Research Agenda for Adaptive Self-Organisation 提出内在动机机制以解决自适应自组织问题 reinforcement learning world model world models
27 Turn-level Multiscale Density Ratio Estimation for LLM Agents 提出Turn-level Multiscale Density Ratio Estimation以解决多轮任务对LLM的挑战 PPO DPO large language model
28 Interactive Memory Learning for Long-Term Conversations 提出交互记忆学习以解决长期对话中的记忆管理问题 reinforcement learning large language model
29 ReDraft, Don't Just Distill: Reference-Driven Revision for Continual VLLM Post-Training 提出ReDraft以解决持续后训练中的能力保持与遗忘问题 distillation multimodal

🔬 支柱三：空间感知与语义 (Perception & Semantics) (2 篇)

#题目一句话要点标签🔗⭐
30 Affect-Prototype Guided Fusion for Open-Vocabulary Incomplete Multi-modal Emotion Recognition 提出情感原型引导融合方法以解决开放词汇不完整多模态情感识别问题 open-vocabulary open vocabulary multimodal
31 Grounding SWE-Agent Decisions in Architecture-0 Design: Navigating Unknown Unknowns through Physical Mapping 提出物理映射保护机制以解决自主软件工程代理的架构设计问题 affordance

⬅️ 返回 cs.AI 首页 · 🏠 返回主页