Model techniques map
Techniquespost-trainingpolicy distillation

implementation detail · filed under post-training

Teacher-Trajectory and SFT-History Reuse

A prefix-conditioning implementation that reuses histories from teacher trajectories and supervised fine-tuning demonstrations.

source
1
model
1
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.

reusing histories from teacher trajectories and SFT demonstrations so decision points train without regenerating preceding turns

usedpost trainingin MiMo-V2.6-Pro-RLXiaomi

Filed alongside

Other methods under post-training :: policy distillation.