implementation detail · filed under training objective
Language modeling loss plus KL loss
Combines a language modeling loss with a KL loss.
Also called LM Loss + KL Loss.
- 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.
LM Loss + KL Loss gives the best balance
usedtraining objectivein MiniMax Sparse AttentionMiniMax
Filed alongside
Other methods under training objective :: auxiliary loss.