LLM代理落地关键在于一致性与边界感知，TRACE用自进化技能库破解可靠性难题，值得关注。
arXiv:2608.22793v1 Announce Type: cross Abstract: Reliable deployment of LLM agents in user-facing products depends not on raw task-solving ability bu…
AI自我进化的边界在哪？这篇论文直面“AI设计AI”是创新还是模仿的终极拷问。
arXiv:2608.17471v1 Announce Type: new Abstract: Recent advances in LLM agents have made them increasingly capable of designing methods for complex AI …
让LLM智能体在动态威胁中自我进化防御，告别手工安全规则，安全研究新范式。
arXiv:2608.12977v1 Announce Type: cross Abstract: The expanding operational capabilities of large language model (LLM) agents introduce sophisticated …
LLM智能体自我进化可能反噬，预承诺门控机制防止技能污染，揭秘新解法。
arXiv:2608.05810v1 Announce Type: new Abstract: Self-evolving agents accumulate capability by distilling reusable skills from their execution trajecto…
无需训练数据即可自我进化的搜索代理，突破传统学习范式，实现自主迭代优化。
arXiv:2601.07055v2 Announce Type: replace Abstract: As high-quality data becomes increasingly difficult to obtain, self-evolution without curated trai…
让AI agent自动复盘工作，把经验沉淀为可复用的"技能"，无需手动触发，几天内自动生成4个合理技能，越用越强。
we built an agent plugin that lets an agent reflects after real work saves the reusable part as a "skill", or improves one it already has. No command,…
用强化学习训练时间管理助手，让AI根据反馈自我进化，实现智能日程规划与高效时间分配。
arXiv:2601.11957v4 Announce Type: replace Abstract: Overlapping calendar invitations force busy professionals to repeatedly decide which meetings to a…
首款专为自我进化LLM智能体设计的评估环境，解构智能体框架演进过程，突破传统任务分数评估局限。
arXiv:2606.17546v1 Announce Type: new Abstract: Self-evolving LLM-based agents improve mainly by changing their agent harness: the structured executio…
多模态AI自我评估存在系统性偏差，一种新发现的“偏好崩溃”现象加剧了模型反馈循环中的错误。
arXiv:2606.16682v1 Announce Type: new Abstract: When AI agents use language models to evaluate their own outputs in a feedback loop, systematic biases…
自进化大模型代理真的忠实于过往经验？首项系统性因果研究揭开真相。
arXiv:2601.22436v3 Announce Type: replace Abstract: Self-evolving large language model (LLM) agents continually improve by accumulating and reusing pa…
AI自我进化新突破：将搜索代码从1260行压减至75行，准确率提升7.7%，token消耗减少4倍。
Article URL: https://github.com/001TMF/harness-forge Comments URL: https://news.ycombinator.com/item?id=48526233 Points: 1 # Comments: 0
用程序化世界模型让LLM智能体在虚拟环境中自我进化，突破静态数据限制。
arXiv:2606.12780v1 Announce Type: cross Abstract: Self-evolving agents are expected to improve through interaction without external supervision, but t…
对比评估与拓扑感知结合，让LLM智能体能自主进化技能库，复杂任务适应性显著提升。
arXiv:2606.13317v1 Announce Type: new Abstract: Skill self-evolution methods for LLM agents aim to turn execution trajectories into reusable skill doc…
终端原生的AI编程助手，能读写代码、执行命令、管理Git，具备持久化记忆与自我进化能力，限免MiMo-V2.5中
MiMoCode 是一个终端原生的 AI 编程助手，基于 OpenCode fork 构建，能读写代码、执行命令、管理 Git，通过持久化记忆系统，在多次会话间保持对你项目的深度理解，并自我进化。@Appinn 上面的内容不重要，重要的是 MiMoCode 内置了一个可以直接使用的免费 MiMo A
从AI包装器迈向自主进化代理，揭秘闭环学习架构实现深度研究与CI/CD自动化。
We are officially transitioning from the era of "AI wrappers" to the era of truly autonomous agentic systems. If you’ve spent any time building with L…
LLM智能体如何在开放世界中自主进化？这篇论文提出OpenSkill框架，让智能体持续学习与适应。
arXiv:2606.06741v1 Announce Type: new Abstract: Self-evolving agents requires adaptation after deployment, but existing approaches assume a usable lea…
自进化LLM智能体通过分布内优化实现持续自我提升，无需人工干预
arXiv:2606.07367v1 Announce Type: new Abstract: Large Language Models (LLMs) have recently emerged as powerful controllers for interactive agents in c…
Anthropic警告Claude AI正以超预期速度自我进化，现已有80%以上的代码由AI自身编写，人类控制力面临挑战。
Article URL: https://www.tomshardware.com/tech-industry/artificial-intelligence/anthropic-says-claude-now-writes-more-than-80-percent-of-its-merged-co…
物理AI智能体自我进化技能+形式化验证，确保安全性与可靠性的前沿方法
arXiv:2606.05395v1 Announce Type: cross Abstract: Reusable robot skills are becoming the basic units through which embodied agents turn open-ended ins…
不写提示词，让AI代理从错误中自我进化——颠覆传统修复流程的新范式。
Imagine deploying an autonomous AI agent to handle your production database migrations, customer support, or code reviews. On day one, it performs bea…