cs.AI（2026-01-07）

📊 共 15 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) (9 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (6)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (9 篇)

#题目一句话要点标签🔗⭐
1 EntroCoT: Enhancing Chain-of-Thought via Adaptive Entropy-Guided Segmentation EntroCoT：通过自适应熵引导分割增强思维链推理 large language model chain-of-thought
2 HoneyTrap: Deceiving Large Language Model Attackers to Honeypot Traps with Resilient Multi-Agent Defense HoneyTrap：利用多智能体防御欺骗大语言模型攻击者，构建蜜罐陷阱 large language model
3 Deontic Knowledge Graphs for Privacy Compliance in Multimodal Disaster Data Sharing 提出基于义务知识图的框架以解决多模态灾害数据共享中的隐私合规问题 multimodal
4 Agent Drift: Quantifying Behavioral Degradation in Multi-Agent LLM Systems Over Extended Interactions 提出Agent Drift概念与ASI指标，量化多Agent LLM系统长期交互中的行为退化问题。 large language model
5 Quantifying the Impact of Modules and Their Interactions in the PSO-X Framework 量化PSO-X框架中模块及其交互对粒子群优化算法性能的影响 multimodal
6 From Laboratory to Real-World Applications: Benchmarking Agentic Code Reasoning at the Repository Level RepoReason：提出仓库级代码推理白盒基准，诊断Agent代码能力 large language model
7 MHRC-Bench: A Multilingual Hardware Repository-Level Code Completion benchmark 提出MHRC-Bench，首个多语言硬件代码仓库级代码补全基准 large language model
8 STAR-S: Improving Safety Alignment through Self-Taught Reasoning on Safety Rules 提出STAR-S框架，通过自学习安全规则推理提升LLM的安全性对齐 large language model ✅
9 Evolving Programmatic Skill Networks 提出程序化技能网络PSN，用于开放环境下的持续技能学习 large language model

🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)

#题目一句话要点标签🔗⭐
10 MobileDreamer: Generative Sketch World Model for GUI Agent MobileDreamer：为GUI代理构建生成式草图世界模型，提升长时任务性能。 world model dreamer
11 Interleaved Tool-Call Reasoning for Protein Function Understanding 提出PFUA：一种交错工具调用的蛋白质功能理解框架，显著提升预测性能。 reinforcement learning large language model chain-of-thought
12 ReEfBench: Quantifying the Reasoning Efficiency of LLMs 提出ReEfBench框架以量化大型语言模型的推理效率 distillation large language model chain-of-thought
13 Anti-Length Shift: Dynamic Outlier Truncation for Training Efficient Reasoning Models 提出动态离群点截断(DOT)方法，解决推理模型训练中的长度偏移问题，提升效率与性能。 reinforcement learning chain-of-thought
14 ROI-Reasoning: Rational Optimization for Inference via Pre-Computation Meta-Cognition 提出ROI-Reasoning，通过预计算元认知优化LLM在预算约束下的推理性能。 reinforcement learning large language model
15 Sandwich Reasoning: An Answer-Reasoning-Answer Approach for Low-Latency Query Correction 提出SandwichR，通过答案-推理-答案范式实现低延迟高精度查询纠错 reinforcement learning chain-of-thought

⬅️ 返回 cs.AI 首页 · 🏠 返回主页