cs.AI（2026-01-26）

📊 共 21 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) (14 🔗3) 支柱二：RL算法与架构 (RL & Architecture) (7 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (14 篇)

#题目一句话要点标签🔗⭐
1 Deconstructing Instruction-Following: A New Benchmark for Granular Evaluation of Large Language Model Instruction Compliance Abilities 提出MOSAIC基准，用于细粒度评估大语言模型指令遵循能力 large language model instruction following
2 RareAlert: Aligning heterogeneous large language model reasoning for early rare disease risk screening RareAlert：对齐异构大语言模型推理，用于罕见病早期风险筛查 large language model
3 Mitigating the OWASP Top 10 For Large Language Models Applications using Intelligent Agents 利用智能代理缓解大语言模型应用中OWASP Top 10安全风险 large language model
4 Think-Augmented Function Calling: Improving LLM Parameter Accuracy Through Embedded Reasoning 提出Think-Augmented Function Calling，通过嵌入式推理提升LLM函数调用参数准确率 large language model chain-of-thought
5 TSRBench: A Comprehensive Multi-task Multi-modal Time Series Reasoning Benchmark for Generalist Models TSRBench：一个综合性的多任务多模态时间序列推理基准，用于评估通用模型。 multimodal ✅
6 A Balanced Neuro-Symbolic Approach for Commonsense Abductive Logic 提出一种平衡神经符号的归纳逻辑方法，提升常识推理能力 large language model
7 Stability as a Liability:Systematic Breakdown of Linguistic Structure in LLMs 揭示LLM训练稳定性与生成质量的矛盾：稳定训练导致语言结构退化 large language model
8 Design Techniques for LLM-Powered Interactive Storytelling: A Case Study of the Dramamancer System Dramamancer：基于LLM的交互式叙事设计技术研究 large language model
9 FadeMem: Biologically-Inspired Forgetting for Efficient Agent Memory 提出FadeMem，一种受生物启发的记忆架构，提升Agent记忆效率并减少存储占用。 large language model
10 FastInsight: Fast and Insightful Retrieval via Fusion Operators for Graph RAG FastInsight：通过融合算子实现图RAG的快速且深刻的检索 large language model
11 A Generative AI-Driven Reliability Layer for Action-Oriented Disaster Resilience Climate RADAR：基于生成式AI的行动导向型灾害韧性可靠性层 large language model
12 TAM-Eval: Evaluating LLMs for Automated Unit Test Maintenance TAM-Eval：用于评估LLM在自动化单元测试维护中的性能的框架与基准 large language model ✅
13 Generative AI in Saudi Arabia: A National Survey of Adoption, Risks, and Public Perceptions 沙特阿拉伯生成式AI应用调查：揭示采用现状、风险认知与公众期望 multimodal
14 EvolVE: Evolutionary Search for LLM-based Verilog Generation and Optimization 提出EvolVE框架以优化LLM驱动的Verilog生成 large language model ✅

🔬 支柱二：RL算法与架构 (RL & Architecture) (7 篇)

#题目一句话要点标签🔗⭐
15 AdaReasoner: Dynamic Tool Orchestration for Iterative Visual Reasoning AdaReasoner：用于迭代视觉推理的动态工具编排框架 reinforcement learning large language model multimodal
16 OffSeeker: Online Reinforcement Learning Is Not All You Need for Deep Research Agents 提出OffSeeker，利用离线数据训练高性能研究Agent，降低在线强化学习成本。 reinforcement learning DPO
17 Emergent Cooperation in Quantum Multi-Agent Reinforcement Learning Using Communication 提出基于通信的量子多智能体强化学习方法，促进序列社会困境中的涌现合作。 reinforcement learning
18 daVinci-Dev: Agent-native Mid-training for Software Engineering daVinci-Dev：面向软件工程的Agent原生Mid-training方法 reinforcement learning large language model
19 AI Agent for Reverse-Engineering Legacy Finite-Difference Code and Translating to Devito 提出AI Agent框架，用于逆向工程传统有限差分代码并转换为Devito。 reinforcement learning large language model
20 ShopSimulator: Evaluating and Exploring RL-Driven LLM Agent for Shopping Assistants ShopSimulator：用于评估和探索强化学习驱动的LLM购物助手 reinforcement learning large language model ✅
21 Diffusion Model-based Reinforcement Learning for Version Age of Information Scheduling: Average and Tail-Risk-Sensitive Control 提出基于扩散模型的强化学习算法，用于优化多用户无线系统中的版本信息年龄调度。 reinforcement learning

⬅️ 返回 cs.AI 首页 · 🏠 返回主页