TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

The Socratic Method for Self-Discovery in Large Language Models

4 点作者 itsuka大约 2 年前

1 comment

cs702大约 2 年前
It&#x27;s a clever idea: The authors instruct multiple LLM-simulated agents to assume different roles in a Socratic dialogue to figure out how to solve problems proposed by the user. Think of it as a <i>team</i> of LLM-simulated agents, each with a different set of goals and responsibilities (e.g., one may be creative, another may be analytical, another may be a proofreader, etc.).<p>Demo (video): <a href="https:&#x2F;&#x2F;vimeo.com&#x2F;821089266" rel="nofollow">https:&#x2F;&#x2F;vimeo.com&#x2F;821089266</a><p>Code: <a href="https:&#x2F;&#x2F;github.com&#x2F;RunzheYang&#x2F;SocraticAI">https:&#x2F;&#x2F;github.com&#x2F;RunzheYang&#x2F;SocraticAI</a>