Open Models Almanac
Methodspost-trainingreinforcement learning algorithm

single mention · post-training

Negative-advantage penalties for malformed reasoning and tool calls

Assigns negative advantage to tokens responsible for malformed reasoning or tool calls.

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

Filed under post-training › reinforcement learning algorithm