3 ms·
It has already been done: http://arxiv.org/abs/2207.05221 http://arxiv.org/abs/2207.05221 However, it seems that RLHF considerably reduces the model's calibrat
by blackbear_ 3y ago
It has already been done: http://arxiv.org/abs/2207.05221 http://arxiv.org/abs/2207.05221
However, it seems that RLHF considerably reduces the model's calibration, so perhaps the method above won't be applicable to ChatGPT and similar.