Model techniques map
Techniquespost-trainingreward modelling

specific method · filed under post-training

Test-difficulty-driven code reward

Uses test difficulty to shape code-task rewards and address sparse rewards on challenging problems.

source
1
models
2
lab adopt it
1
strongest
used

How sources treat it

One count per evidence span, weakest treatment to strongest.

used 1

Documented in

Evidence

1 span quoted from the sources, strongest treatment first.

To mitigate the sparse reward issue for challenging code problems, we introduce a test difficulty driven code reward.

usedtraining objectivein MiMo-7B-RLXiaomi

Filed alongside

Other methods under post-training :: reward modelling.