Open Models Almanac
Methodspost-trainingreward modelling

single mention · post-training

Sparse terminal-verifier reward

Uses a binary verifier at trajectory end and propagates its reward through token-level advantages.

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

Also called binary verifier.

Filed under post-training › reward modelling