Model techniques map
Techniquespost-trainingreinforcement learning algorithm

implementation detail · filed under post-training

Dropping All-Zero-Advantage Groups

A training implementation that drops groups whose advantages are all zero.

source
1
models
2
lab adopt it
1
strongest
default

How sources treat it

One count per evidence span, weakest treatment to strongest.

default 1

Documented in

Evidence

1 span quoted from the sources, strongest treatment first.

groups whose advantages are all zero are dropped by default

defaultunclearin MiMo-V2.6 RL and OPD infrastructureXiaomi

Filed alongside

Other methods under post-training :: reinforcement learning algorithm.