Model techniques map
Techniquespost-trainingreward modelling

specific method · filed under post-training

Groupwise Reward Synthesis

Builds task-specific rubrics from contrasting offline rollouts and combines rubric-based quality scores with test outcomes.

Also called Groupwise Reward Synthesis (GRS).

sources
3
models
2
lab adopt it
1
strongest
core

How sources treat it

One count per evidence span, weakest treatment to strongest.

used 3core 1

Documented in

Evidence

4 spans quoted from the sources, strongest treatment first.

Groupwise Reward Synthesis (GRS) builds task-specific rubrics offline from contrasting rollouts and fuses rubric quality with test outcomes

coretraining objectivein MiMo-V2.6-Flash-RLXiaomi

Groupwise Reward Synthesis (GRS) builds task-specific rubrics offline from contrasting rollouts and fuses rubric quality with test outcomes

usedtraining objectivein MiMo-V2.6-Pro-RLXiaomi

Through our proposed Groupwise Reward Synthesis (GRS) and Groupwise Advantage Redistribution (GAR), these fine-grained distinctions are converted into more informative learning signals

usedtraining objectivein MiMo-V2.6 RL trainingXiaomi

It compares multiple offline rollouts to construct task-specific rubrics, which are reused during training to score individual rollouts and combine their quality scores with test rewards.

usedpost trainingin MiMo-V2.6-Pro and MiMo-V2.6-FlashXiaomi

Filed alongside

Other methods under post-training :: reward modelling.