LeCun：纯 LLM 才会做自回归 token 预测，好代码生成器不是

精选理由

LeCun 发推澄清：别把代码生成器和纯 LLM 自回归划等号，观点直接，适合关注大模型技术路线的人。

AI 摘要

Yann LeCun 在 X 平台回应网友评论，澄清他批评的对象是纯 LLM 所采用的自回归 token 预测。他表示，优秀的代码生成系统并不属于纯 LLM，工作方式也不是简单的自回归 token 预测。这条推文获得 133 次点赞、25 条回复，浏览量超过 1.7 万。

原文 · Yann LeCun

@gabriberton You obviously did not understand my statement. I was talking about auto-regressive tok...

@gabriberton You obviously did not understand my statement. I was talking about auto-regressive token prediction, which is what pure LLMs do. But good code generation systems aren't pure LLMs and aren't doing mere auto-regressive token prediction. 💬 25 🔄 5 ❤️ 133 👀 17334 📊 33 ⚡

查看原推