cs.AI（2026-10-21）

📊 共 24 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) (15 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (8 🔗1) 支柱一：机器人控制 (Robot Control) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (15 篇)

#	题目	一句话要点	标签	🔗	⭐
1	Automated Materials Data and Knowledge Extraction using Machine Learning and Large Language Models	利用机器学习和大语言模型实现材料数据与知识的自动提取	large language model
2	Same Answer, Different Reasons: Trace-Level Divergence in Multilingual Large Language Models	提出追踪级别差异框架以解决多语言模型可解释性问题	large language model
3	Lineage-Aware Search Control for LLM-Based Evolutionary Factor Mining	提出LineageEvo以解决LLM基础的因子挖掘搜索控制问题	large language model
4	Commit or Abstain? System One Decision Models, LLMs, and Missing Legal Authority	提出基于系统一决策模型的法律AI以解决法律权威验证问题	large language model
5	ToM2Vec: Continuous Theory-of-Mind Tokens for Multi-Agent Coordination	提出ToM2Vec以解决多智能体协作中的理论心智建模问题	large language model
6	Metamodel-Guided LLM Decoding for Consistent DSL Generation	提出MeMoDecode以解决DSL生成中的一致性问题	large language model
7	From Prompts to Posteriors: Can LLM Agents Perform Bayesian Inference in Context?	评估大型语言模型在上下文中进行贝叶斯推断的能力	large language model
8	Same Probabilities, Different Advice: Gain-Loss Framing Effects in High-Stakes LLM Decision Support	提出FrameBench-HPS以解决高风险决策中的框架效应问题	large language model
9	ARCANA: A Benchmark for Abstraction and Analogical Reasoning	提出ARCANA基准以解决远类比推理问题	large language model	✅
10	AI-Based End-to-End Automotive Design and Software Development for SDVs	提出基于AI的工具链以实现软件定义车辆的自动化设计与开发	large language model
11	ORSMT: Unleashing the Potential of Satisfiability Modulo Theories for Semantic Verification and Repair in LLM-Based Optimization	提出ORSMT以解决大语言模型优化中的语义验证与修复问题	large language model
12	Agentic Music Post-Production	提出MixAgent以解决音乐后期制作中的混音质量问题	multimodal
13	Does Every Instance Count Equally? Order Effects in LLM Multi-Instance Processing	研究实例顺序对LLM多实例处理的影响	large language model
14	AutoFraming: Synthesizing Dynamic Frames of C Code with LLMs	提出AutoFraming以解决C代码中帧条件合成不足的问题	large language model
15	Inheritance as an Admission Test for LLM Prompt Merge Operators	提出基于继承的LLM合并操作的入学测试方法	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	🔗	⭐
16	Steering Black-Box Large Language Model Agents with Pitfall Anticipation World Model	提出陷阱预测世界模型以提升黑箱大语言模型代理的表现	world model world models large language model
17	LLM-Based Exogenous World Modeling for Dynamic Scheduling	提出基于LLM的外部世界建模以解决动态调度问题	world model world models large language model
18	Parallel Thinking for Large Language Model with Path Diversity	提出多样化并行思维方法以解决推理路径同质化问题	reinforcement learning large language model	✅
19	Events, Not Pixels: Where World Models Get the Return Wrong	提出事件过渡模型以解决世界模型回报错误问题	world model world models
20	TeachBack: Learning to Teach via Bidirectional Preference Optimization for On-Policy Self-Distillation	提出TeachBack框架以优化在政策自蒸馏中的教师学习方式	preference learning distillation privileged information
21	An Evidence-Driven Multi-Agent Compound AI System for Head and Neck Proton Radiotherapy Planning	提出基于多智能体复合AI系统的头颈部质子放疗计划优化方法	DRL PPO instruction following
22	Conflicted Records: Typed Adversarial Contracts for Auditable LLM Output Quality Enforcement	提出BiasGuard-D以解决LLM输出质量审计问题	RLHF chain-of-thought
23	RubricOpt: Self-Revision of Semantic Rubrics for Preference Optimization	提出RubricOpt以优化偏好学习中的语义标准	DPO direct preference optimization

🔬 支柱一：机器人控制 (Robot Control) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
24	From Association to Decision Value: Testing Paired Instruction Dependence for VLA Corpus Selection	提出基于配对指令依赖的决策值评估方法以优化机器人示范选择	quadruped VLA

⬅️ 返回 cs.AI 首页 · 🏠 返回主页