koi691337

joined 11 months ago
sorted by: new top controversial old
[R] Orca 2: Teaching Small Language Models How to Reason in c/machinelearning@academy.garden
[–] koi691337@alien.top 1 points 11 months ago (1 children)

Then you could have the language model generate imagined user responses and optimize the reward signal on the imagined user responses

Wouldn't this just constitute to the model sort of overfitting to noise?