揭秘工具架构设计如何影响LLM Agent后训练效果，为智能体优化提供新视角
arXiv:2606.25447v1 Announce Type: new Abstract: Tool-integrated LLM agents are often wrapped within a harness: the scaffolding that determines which t…
研究表明，多LLM系统的对话行为更多取决于后训练配方，而非模型家族差异，挑战了传统选型策略。
arXiv:2606.20632v1 Announce Type: cross Abstract: Multi-LLM systems use multiple language models to deliberate, judge each other's outputs, or coordin…
重新定义大模型后训练本质，揭示其等同于大规模监督学习的关键理论，打开训练范式新视角。
arXiv:2606.07527v1 Announce Type: cross Abstract: The prevailing paradigm for training LLMs has evolved to rely on a massive post-training phase consi…
Mercor AI 正在招募通才专家参与大模型后训练，内推链接直达岗位详情。
Article URL: https://work.mercor.com/jobs/list_AAABnd0_i7HkvuPZUt1LkbpI/generalist-expert Comments URL: https://news.ycombinator.com/item?id=48250861 …