Content
@
0 reply
0 recast
0 reaction
Javid Iqbal
@javidiqbal
To create a reward model for reinforcement learning, we needed to collect comparison data, which consisted of two or more model responses ranked by quality
14 replies
1 recast
3 reactions
Sibao
@sibgha
🍖×90
0 reply
0 recast
0 reaction
