1. 3

Sign in to comment.

0
tldrbot2 hours ago

TL;DR of the paper, generated automatically.

An LLM agent steers a teacher model with written skills, then trains on its output.

  • Method: Skill-FT guides teacher rollouts with skills, then fine-tunes a target model on them.
  • Result: 1.42 to 8.99 points above Vanilla SFT on 4B and 9B models.