cs.AI（2026-10-30）

📊 共 45 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (30 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (15)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (30 篇)

#题目一句话要点标签🔗⭐
1 Sensorium: An Open-Source Multimodal Data Collection and Curation Library for Embodied Surgical AI 提出Sensorium以解决多模态外科数据收集难题 multimodal
2 With foundation-model features, aggregator complexity does not improve cross-hospital discrimination in gastric whole-slide image classification 提出CHIME-MIL以解决跨医院胃癌全切片图像分类问题 foundation model
3 ID2Pattern: Pretraining Generalizable Identity Behavior Patterns for Transaction Risk Foundation Models 提出ID2Pattern以解决在线支付系统身份行为模式识别问题 foundation model
4 Prune Gradually, Adapt Sparsely: Efficient Pruning of Large Language Models at Extreme Sparsity 提出ETAPS以解决大语言模型高稀疏性下的质量保持问题 large language model
5 General-purpose foundation models surpass specialized ones on chemical property prediction tasks 通用基础模型在化学性质预测任务中超越专业模型 foundation model
6 RECAST: REconstruction of Chain-of-Thought via Auditable Structured Traces 提出RECAST框架以重建闭源模型的推理过程 chain-of-thought
7 Advancing Clinical Trial Outcome Prediction with Domain-Specific Large Language Models 提出领域特定的大语言模型以提升临床试验结果预测 large language model
8 Specification-Guided Generation for Large Language Models 提出神经符号框架以解决大语言模型的规范满足问题 large language model
9 Do Transaction Prices Reflect Visual Preference? A Diagnostic Study with Multimodal Language Models 提出MarketImage框架以解决交易价格与视觉偏好识别问题 multimodal
10 Experts Exhibit Structured Redundancy: Structure-Preserving Expert Pruning for Mixture-of-Experts Large Language Models 提出结构保留专家剪枝方法以优化MoE大语言模型 large language model
11 QUAL: Quantile Alignment Improves Zero-Shot Time-Series Foundation Models 提出QUAL以改善零-shot时间序列基础模型的预测准确性 foundation model ✅
12 TradeWar: A Benchmark for Execution-Grounded Trade Negotiation with Adaptive LLM Agents 提出TradeWar基准以解决执行导向的贸易谈判问题 large language model chain-of-thought
13 ImplicitHop: Evaluating knowledge integration across context and memory 提出ImplicitHop以评估知识在上下文和记忆中的整合能力 large language model chain-of-thought
14 From the Zoo to the BULLPEN: Sample-Efficient LLM Profiling via Multi-Channel Model Embeddings 提出BULLPEN框架以高效评估大型语言模型的行为特征 large language model ✅
15 ACCT: Asymmetric Cross-LLM Context Transfer 提出ACCT以解决跨模型上下文重用的计算成本问题 large language model
16 ReX-GG: a LLM Ensemble Pipeline for Relation-extraction and Graph Generation 提出ReX-GG以解决关系提取与图生成中的模型脆弱性问题 large language model
17 Automated Keyword Generation Using LLMs in a High-Volume Data Virtualization Service 提出基于大语言模型的自动关键词生成方法以优化数据虚拟化服务 large language model
18 AI evaluators favor job applications written by certain AI models, independent of candidate quality 提出多模型审计框架以解决AI招聘筛选中的偏见问题 large language model
19 Breaking Information Islands in Sparse Tuning via Small-World Connectivity 提出SWIFT以解决稀疏调优中的信息孤岛问题 large language model ✅
20 Real Computation or Merely Naming? A Pre-Registered Mechanism Audit of 25 Cognitive-Architecture Modules in an LLM-Based Agent System 提出四问审计协议以验证LLM代理系统的认知模块机制 large language model
21 Multi-Test Nation Bias Benchmarking of LLMs with Explicit Unbiased Baselines 提出多测试国家偏见基准以评估大型语言模型的偏见问题 large language model
22 When Safety Checks Become Privacy Risks: A Privacy-Preserving Paradigm for LLM Agent Safety Auditing 提出隐私保护审计范式以解决LLM代理安全性问题 large language model
23 SERIAL-7: Deriving, Realizing, and Preserving Machine-Verifiable Outputs Across Representations 提出SERIAL-7以解决机器可验证输出生成问题 large language model
24 Self-Improvement with Auxiliary Rewards: Learning from Verified Traces and Evaluating Transfer 提出辅助奖励机制以提升小语言模型自我改进能力 instruction following
25 TRIDENT and SHAKTI: A Framework and Benchmark for Hindi AI-Text Detection 提出SHAKTI基准与TRIDENT框架以解决印地语AI文本检测问题 large language model
26 ROSTrans: Empirically Benchmarking LLMs for ROS1-to-ROS2 Migration 提出ROSTrans以评估大语言模型在ROS1到ROS2迁移中的表现 large language model
27 LLM Detoxification via Transition-Aware Reward-Guided Representation Sampling 提出TARGRS以解决大语言模型生成有毒内容的问题 large language model
28 When Disclosure Is Compulsory: Evaluating Ride-Hailing Rebalancing Policies with Self-Interested LLM Driver Agents 利用自利型LLM代理评估强制披露下的网约车再平衡政策 large language model
29 Calibration of Large Model Quantization based on HiFloat4 Format 提出基于HiFloat4格式的模型量化校准方法以提升推理精度 large language model
30 CIRCE Divines LLM Task Complexity from Constraint Structure 提出CIRCE以评估LLM任务复杂性 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (15 篇)

#题目一句话要点标签🔗⭐
31 scAnyOmics: Foundation-Model Distillation for Any-to-Any Completion of Multi-omics Virtual Cells 提出scAnyOmics以解决多组学虚拟细胞信息缺失问题 distillation foundation model multimodal
32 Multimodal In-Context Learning with Explicit Rule Induction 提出ERIM框架以解决多模态上下文学习中的规则推导问题 reinforcement learning large language model multimodal
33 Erase the Injected Region: Multimodal Prompt-Injection Defense with On-Policy Self-Distillation 提出ERASE以解决多模态提示注入攻击问题 distillation multimodal
34 CausalWAM: Learning the Causes of Actions in World Action Models 提出CausalWAM以解决世界动作模型中的因果关系问题 world model world models world action model
35 One Verdict Is Not Enough: Regression Testing Learned World Models and Their Test Harnesses 提出TetraState以解决学习世界模型回归测试中的不确定性问题 world model world models JEPA
36 When Does Test-Time Search Help World Action Models? An Empirical Audit of Candidate Selection and Simulator Integrity 审计测试时搜索对世界行动模型的影响以优化控制策略 world action model world action models
37 RTL-Interact: Training EDA-Interactive Models for Multi-Task RTL Design 提出RTL-Interact以解决RTL设计中的EDA交互问题 reinforcement learning large language model chain-of-thought
38 Chance-Constrained Inspection Viewpoint Sequencing for On-Orbit Servicing via Safe Reinforcement Learning 提出基于机会约束的视角序列规划以解决在轨服务问题 reinforcement learning PULSE
39 DAMI: Dual-Axis Memory Initialization for Enhanced Long-Context Retrieval in Gated Linear Attention 提出DAMI以解决长上下文检索中的记忆初始化问题 Mamba linear attention
40 ART of PIV: Agentic Real-Time Particle Image Velocimetry for Active Flow Control 提出ART of PIV以解决实时粒子图像测速中的配置选择问题 reinforcement learning dreamer optical flow
41 Structure-Aware Skill Retrieval and Arbitration for Optimization Modeling 提出ArchOR以解决优化建模中的技能检索与仲裁问题 representation learning large language model
42 Risk-Sensitive Mission Planning for Multi-Target On-Orbit Inspection via Distributional Reinforcement Learning 提出基于分布式强化学习的风险敏感多目标轨道检查任务规划方法 reinforcement learning
43 What Should LLMs Forget: Reinforcement Unlearning of Knowledge Associations 提出知识关联强化遗忘方法以解决LLMs知识遗忘问题 reinforcement learning large language model
44 Fuzzy selective state model for forecasting the dynamics of consumer preferences 提出模糊选择状态模型以预测消费者偏好的动态变化 Mamba state space model
45 Dynamical Memory Transformer: Selective Recurrent Memories with Constant Cost Attention 提出动态记忆变换器以解决长序列信息检索问题 Mamba state space model

⬅️ 返回 cs.AI 首页 · 🏠 返回主页