首页 > AI前沿 > Learning through Discernment: when a small language model learns from user feedback on the job, and when it decays

Learning through Discernment: when a small language model learns from user feedback on the job, and when it decays

arXiv自然语言 2026-11-18 10:18 20 阅读 查看原文

Small language models on devices could keep improving by learning, overnight, from their own daily work. Without a true outcome, however, something must decide which attempts are worth learning from, and both the model's own judgment and the user's reaction can be wrong.

We study a small model (Qwen3 0.6B and 1.7B) with a LoRA adapter, fine-tuned each night on the attempts a judge accepts, over 127 lives of up to a year on WebShop and $τ