cs.AI（2026-10-19）

📊 共 41 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (24 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (16 🔗3) 支柱三：空间感知与语义 (Perception & Semantics) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (24 篇)

#题目一句话要点标签🔗⭐
1 Toward Structured Financial Foundation Models 提出结构化金融基础模型以应对金融系统挑战 foundation model
2 Beyond Static Leaderboards: Capability-Focused Evaluation with Structural Diversity for Time Series Foundation Models 提出CaFE以解决时间序列基础模型评估的局限性问题 foundation model
3 Towards More Granular Reasoning: Mapping the DSA Transparency Database Taxonomy to Platform Terms and Conditions Using Large Language Models 通过大语言模型映射DSA透明数据库分类以提升内容审核透明度 large language model
4 MoveBench: A Rubric-Based Benchmark for Evaluating Large Language Models in Sports and Health Coaching 提出MoveBench以解决体育与健康领域LLM评估缺乏标准的问题 large language model ✅
5 Multimodal Decoupling for Diversified Recommendation through Semantic and Structural Shaping 提出MOSS框架以解决推荐系统的准确性与多样性权衡问题 multimodal
6 EvalRx: Agentic Diagnosis and Verified Repair of Foundation Models 提出EvalRx框架以解决模型诊断与修复问题 foundation model
7 PFedQFT: Personalized Federated Quantum-Inspired Fine-Tuning for Medical Large Language Models 提出PFedQFT以解决医疗数据隐私与适应性问题 large language model
8 LIRQuant: An Efficient Locally Isolated Rotation Quantization Method for Large Language Models 提出LIRQuant以解决大语言模型低位量化中的激活异常问题 large language model
9 Beyond Leaderboards: A Method for Application-Specific Capability Profiling of Large Language Models 提出应用特定能力分析方法以优化大型语言模型选择 large language model
10 Reasoning-Aware Multicalibrated Reliability Predictors for Language Models 提出推理感知的多校准可靠性预测器以提升语言模型的可靠性 large language model chain-of-thought
11 WhatIf-Bench: Benchmarking Intervention-Aware Time Series Forecasting 提出WhatIf-Bench以解决干预感知时间序列预测问题 foundation model
12 RouterWorld: Scaling Conversational AI with Latency-Aware LLM Routing 提出RouterWorld以解决对话AI中的延迟感知LLM路由问题 large language model
13 Are Self-Evolving Agents Continual Learners? 提出自演化代理以解决持续学习问题 large language model
14 Honeypots for AI Agent Insider Threats 提出蜜罐技术以应对AI代理的内部威胁 chain-of-thought
15 Autonomous intelligence for mining productivity: a multi-agent AI approach to operational value creation 提出EI2ROC以解决矿业IROCs自主性不足问题 large language model
16 GAIDA: Generative AI for Human-in-the-Loop Document Annotation 提出GAIDA以优化文档标注过程中的人机协作 large language model
17 TimesJev: Thinking Fast on Time Series with System 1 Sufficiency 提出TimesJev以解决时间序列推理中的延迟问题 large language model
18 Procedural Memory for Language Agents: A Neurosymbolic Perspective on Explicit Goal Structure 提出目标导向图以解决语言模型的程序性记忆缺失问题 large language model
19 GLOBE: Geometry-Led Online Behavioral Early Exit via Unit-Sphere Trajectories 提出GLOBE以解决链式思维推理过程中的冗余生成问题 chain-of-thought
20 A Cue's Hidden Price: Auditing Seller Influence on AI Shoppers 提出价格等价审计以量化卖家对AI购物者的影响 large language model
21 Standing on the Shoulders of Giants: Enhancing LLM Agents for Automated Fact-Checking with Human Expert Skills 提出PFC-Skill以提升LLM代理的自动化事实核查能力 large language model
22 From Black Box to Controller: Inference-Time Steering of LLM via Feature Amplification 提出NeuroSteer以解决LLM推理时行为调整问题 large language model
23 CVaR-RMD: CVaR Selection Pressure for Robust Metaheuristic Discovery 提出CVaR-RMD以解决启发式算法选择中的风险敏感性问题 large language model
24 When Safety Fails Collectively: Robust Alignment Against Harmful Fine-Tuning via Joint Weight Perturbations 提出联合权重扰动以应对有害微调问题 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (16 篇)

#题目一句话要点标签🔗⭐
25 CODA: Cohort- and Drift-aware Foundation Model for Multimodal Clinical Reasoning 提出CODA以解决临床推理中的群体与个体差异问题 reinforcement learning foundation model multimodal
26 Think, Compress, Answer: Reinforcement Learning for Multimodal Grounding via Reasoning Compression 提出TCA框架以解决多模态定位中的推理压缩问题 reinforcement learning large language model multimodal
27 EmbodiedWorld-200K: Toward Embodied World Agents via Embodied-Planning Dataset and Interactive World Models 提出EmbodiedWorld-200K数据集以推动开放世界的具身智能体研究 reinforcement learning world model world models ✅
28 TRACE: Tracking Multimodal Clinical Trajectories for Inpatient Risk Prediction 提出TRACE以解决住院患者风险预测中的多模态临床轨迹建模问题 contrastive learning multimodal
29 Agent World Model in the Wild: Computer-Use Agent Environment Synthesis 提出WildAWM以合成可控环境解决计算机使用代理训练问题 world model world models ✅
30 AnyGoal: On World Model Generalization Beyond Expert Rollouts 提出AnyGoal以解决世界模型在目标转移任务中的泛化问题 world model world models
31 ActTime: Action-Conditioned World Modeling for Time Series 提出ActTime以解决时间序列中的行动条件建模问题 world model world models
32 Productive State-Space Exploration for Multi-turn RL in LLM Agents 提出条件状态空间探索框架以解决多轮RL中的低效行为问题 reinforcement learning policy learning large language model
33 ToolOPD: Credit-Aware On-Policy Distillation for Tool-Use Agents 提出ToolOPD以解决长时间工具使用中的信用分配问题 reinforcement learning distillation
34 Training Natural Schemers: RL Can Amplify Oversight-Conditional Behavior 提出自然模型以研究AI在监督条件下的策划行为 reinforcement learning chain-of-thought
35 VeriCUA: Learning Proactive Verification for Reliable GUI Agent Rewards 提出VeriCUA以解决GUI代理奖励验证的可靠性问题 reinforcement learning foundation model
36 Train An Imaginative Agent 提出Agentic Imagination框架以解决Sokoban等环境中的规划问题 reinforcement learning world model world models
37 Tandem: Serving Hybrid Linear-Attention GUI Agents on Mobile NPUs 提出Tandem以提升移动NPU上混合线性注意力GUI代理的性能 linear attention
38 Selective Transfer through Sequence Distillation: From Benchmark Knowledge to Backdoor Behaviour 提出选择性蒸馏方法以解决知识转移中的后门行为问题 distillation ✅
39 Not Every Rollout Needs Distillation: Correctness-Filtered On-Policy Self-Distillation 提出正确性过滤的在线自蒸馏方法以提高计算效率 distillation
40 ClueFlow: Learning Long-Horizon Search through Attributed Research-State Transitions 提出ClueFlow以解决长时间搜索中的状态转移问题 reinforcement learning PPO

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

#题目一句话要点标签🔗⭐
41 SAR: Learning Global Item-Space Matching and Geometry-Aware Placement Ranking for Buffered Online 3D Bin Packing 提出SAR框架以解决在线3D箱装问题 affordance

⬅️ 返回 cs.AI 首页 · 🏠 返回主页