English(EN) A multi-agent LLM where each agent learns when to defer to a human, trained with GRPO on a cost-aware reward. Each defer event becomes SFT data, so the model gr

多智能体LLM学习使用GRPO求助于人类

作者 PulseAugur 编辑部 · [1 个来源] · 2026-05-21 07:01

研究人员开发了一种多智能体大型语言模型，该模型能够学习何时求助于人类输入。该模型使用GRPO在考虑成本的奖励系统上进行训练，并且每次求助都会被用作监督微调数据。这使得模型能够逐步整合人类的专业知识，并且可调的成本参数允许在部署期间在准确性和人工干预预算之间进行权衡。 AI

影响引入了一种新颖的多智能体LLM训练方法，实现了与人类专家自适应协作。

排序理由该集群描述了一篇新颖的研究论文，详细介绍了一种训练多智能体LLM的新方法。[lever_c_demoted from research: ic=1 ai=1.0]

在 Mastodon — fosstodon.org 阅读 →

GRPO
LLM

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 0 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 该集群描述了一篇新颖的研究论文，详细介绍了一种训练多智能体LLM的新方法。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, model release; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: 141 days old; Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] · 2026-05-21 07:01

一种多智能体LLM，每个智能体学习何时向人类寻求帮助，并使用GRPO在成本感知奖励上进行训练。每次寻求帮助的事件都成为SFT数据，因此模型能够学习

A multi-agent LLM where each agent learns when to defer to a human, trained with GRPO on a cost-aware reward. Each defer event becomes SFT data, so the model gradually absorbs the human's expertise. Tunable cost knob trades accuracy against human-call budget at deployment, no ret…

链接 benjaminhan.net/…/20260520-adaptive-colla…

本文如何被排名

报道来源 [1]

一种多智能体LLM，每个智能体学习何时向人类寻求帮助，并使用GRPO在成本感知奖励上进行训练。每次寻求帮助的事件都成为SFT数据，因此模型能够学习

相关实体

相关话题