Distillation depends on a realistic prompt distribution

John Schulman says a realistic prompt distribution can help a distilled model match a larger model. He contrasts this with easily verifiable tasks, which may produce strong benchmark results but weaker performance across a broader distribution.

John Schulman·已发布 2026年9月11日

观点背后的信息

译文仅辅助阅读；核查观点请以原始摘录为准。

AI researchers debate how close we are to recursive self-improvement

原始摘录

If you have a really good realistic prompt distribution for distillation, you can match the big model really well. But if you only have this distribution of easily verifiable tasks, then you can match the big model on all the benchmarks, but you do worse on this broader distribution

此片段未附更多上下文，请阅读原始访谈。

打开原始内容 · 26:48 查看这场访谈

打开该集并跳转至26:48。

时间点来自所提供的转录稿，尚待媒体回放核对。