AI论文速递 2026年05月15日（HuggingFace Daily Papers）¶
数据来源：https://huggingface.co/papers 采集时间：2026-05-15
📌 重点关注¶
- Orchard: An Open-Source Agentic Modeling Framework | arXiv — 【重点关注】 Agentic modeling aims to transform LLMs into autonomous agents capable of sol... 💡 开源代理建模框架，推动LLM向自主代理转变
- PersonalAI 2.0: Enhancing knowledge graph traversal/retrieval with planning mechanism for Personalized LLM Agents | arXiv — 【重点关注】 We introduce PersonalAI 2.0 (PAI-2), a novel framework, designed to enhance l... 💡 知识图谱+规划机制，增强个性化代理检索能力
- Beyond Individual Intelligence: Surveying Collaboration, Failure Attribution, and Self-Evolution in LLM-based Multi-Agent Systems | arXiv — 【重点关注】 LLM-based autonomous agents have demonstrated strong capabilities in reasonin... 💡 多智能体协作进化，AI系统重要研究方向
📋 其他值得关注¶
- An Empirical Study of Automating Agent Evaluation | arXiv — Agent evaluation requires assessing complex multi-step behaviors involving to...
- ATLAS: Agentic or Latent Visual Reasoning? One Word is Enough for Both | arXiv — Visual reasoning, often interleaved with intermediate visual states, has emer...
- Towards Self-Evolving Agentic Literature Retrieval | arXiv — As large language models reshape scientific research, literature retrieval fa...
- EvolveMem:Self-Evolving Memory Architecture via AutoResearch for LLM Agents | arXiv — Long-term memory is essential for LLM agents that operate across multiple ses...
- Self-Distilled Agentic Reinforcement Learning | arXiv — Reinforcement learning (RL) has emerged as a central paradigm for post-traini...
- SPIN: Structural LLM Planning via Iterative Navigation for Industrial Tasks | arXiv — Industrial LLM agent systems often separate planning from execution, yet LLM ...
- Source or It Didn't Happen: A Multi-Agent Framework for Citation Hallucination Detection | arXiv — Large language models are increasingly used in scientific writing, yet they c...