Open Models Almanac
Methodspost-trainingsupervised fine-tuning

specific method · filed under post-training

Self-correction cold start

Creates a cold-start dataset by having the model reflect on and rewrite its own misaligned turns into grounded next steps.

The sources on this page only mention it in passing. The line above is the extraction's one-line gloss. See supervised fine-tuning for the methods filed beside it.

Also called Cold start from self-correction.

sources
2
models
2
lab adopt it
1
strongest
used

Documented in

Evidence

2 quotes from 2 sources, one card per model, each opening with its strongest.

How sources treat it
used 2
↑