Open Models Almanac
Methodspost-trainingagentic post-training

single mention · post-training

Common agent-environment interaction loop

A shared task loop in which an agent acts until its final response, a verifier scores the resulting state, and a hook ends or continues the episode.

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

Also called Agentic tasks share a common loop.

Filed under post-training › agentic post-training