cs.AI（2026-02-11）

📊 共 18 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (11) 支柱二：RL算法与架构 (RL & Architecture) (5 🔗1) 支柱一：机器人控制 (Robot Control) (2)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (11 篇)

#题目一句话要点标签🔗⭐
1 Reinforcing Chain-of-Thought Reasoning with Self-Evolving Rubrics 提出RLCER，利用自进化规则增强思维链推理，无需人工标注。 chain-of-thought
2 OmniSapiens: A Foundation Model for Social Behavior Processing via Heterogeneity-Aware Relative Policy Optimization 提出HARPO算法，训练社交行为处理基础模型Omnisapiens，提升异构数据学习能力。 foundation model
3 Abstraction Generation for Generalized Planning with Pretrained Large Language Models 利用预训练大语言模型为通用规划生成抽象表示 large language model
4 GameDevBench: Evaluating Agentic Capabilities Through Game Development GameDevBench：通过游戏开发评估Agent能力的多模态基准测试 multimodal
5 FeatureBench: Benchmarking Agentic Coding for Complex Feature Development FeatureBench：面向复杂特性开发的Agentic Coding基准测试 large language model
6 Can LLMs Cook Jamaican Couscous? A Study of Cultural Novelty in Recipe Generation 研究表明大型语言模型在食谱生成中未能有效进行文化适应，与人类表现存在显著差异。 large language model
7 Blind Gods and Broken Screens: Architecting a Secure, Intent-Centric Mobile Agent Operating System 提出Aura：一种面向意图的安全移动Agent操作系统架构，解决现有“屏幕即接口”模式的安全漏洞。 large language model
8 VulReaD: Knowledge-Graph-guided Software Vulnerability Reasoning and Detection VulReaD：基于知识图谱的软件漏洞推理与检测方法 large language model
9 The Neurosymbolic Frontier of Nonuniform Ellipticity: Formalizing Sharp Schauder Theory via Topos-Theoretic Reasoning Models 神经符号方法求解非一致椭圆问题：基于拓扑斯理论的形式化绍德理论 chain-of-thought
10 To Think or Not To Think, That is The Question for Large Reasoning Models in Theory of Mind Tasks 研究表明大型推理模型在心理理论任务中表现不稳定，需开发专用能力。 large language model
11 MERIT Feedback Elicits Better Bargaining in LLM Negotiators 提出基于效用反馈的框架，提升LLM谈判者在复杂场景下的议价能力 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
12 Neuro-symbolic Action Masking for Deep Reinforcement Learning 提出神经符号动作掩码(NSAM)，提升DRL样本效率并减少约束违背。 reinforcement learning deep reinforcement learning DRL
13 Found-RL: foundation model-enhanced reinforcement learning for autonomous driving Found-RL：利用基础模型增强自动驾驶强化学习，解决样本效率和可解释性问题 reinforcement learning reward shaping foundation model ✅
14 Interactive LLM-assisted Curriculum Learning for Multi-Task Evolutionary Policy Search 提出交互式LLM辅助的课程学习框架，用于多任务进化策略搜索。 curriculum learning embodied AI multimodal
15 Cross-Sectional Asset Retrieval via Future-Aligned Soft Contrastive Learning 提出FASCL框架，通过未来收益相关性进行横截面资产检索。 representation learning contrastive learning
16 Fine-Tuning GPT-5 for GPU Kernel Generation 通过强化学习微调GPT-5，显著提升GPU Kernel代码生成质量与效率 reinforcement learning large language model

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
17 See, Plan, Snap: Evaluating Multimodal GUI Agents in Scratch 提出 ScratchWorld 基准测试，评估多模态 GUI 智能体在 Scratch 中的编程能力 manipulation multimodal
18 Protecting Context and Prompts: Deterministic Security for Non-Deterministic AI 提出认证提示与认证上下文，实现大语言模型应用中确定性的安全防护。 manipulation large language model

⬅️ 返回 cs.AI 首页 · 🏠 返回主页