implementation detail · filed under training objective
Temperature-scaled forward KL distillation
Uses a temperature-scaled forward-KL loss against the backbone’s logits.
Also called temperature-scaled forward-KL loss.
- sources
- 2
- model
- 1
- lab adopt it
- 1
- strongest
- used
How sources treat it
One count per evidence span, weakest treatment to strongest.
used 2
Documented in
Evidence
2 spans quoted from the sources, strongest treatment first.
We use a temperature-scaled forward-KL loss against the backbone’s logits.
usedunclearin Nemotron 3 UltraNVIDIA
We use a temperature-scaled forward-KL loss against the backbone’s logits.
usedtraining objectivein Nemotron 3 UltraNVIDIA
Filed alongside
Other methods under training objective :: distillation objective.