AdviSD: Learning to Advise Frontier LLMs via Targeted Multi-Turn Self-Distillation
A small trainable advisor can steer a frozen language-model executor using natural-language advice. In addition to learning from task rewards, the advisor can use feedback from completed interactions to improve its advice. However, a plausible correction need not change execution, yet learning from such corrections can still affect the advisor's future decisions in other contexts. In a…
We haven't written up this one. arXiv cs.AI has the full story — the link below goes straight to it.
