Generalized on-policy distillation with reward extrapolation (arxiv.org) 3 points by fzliu 5mo ago ↗ HN
0 comments
[ 4.9 ms ] story [ 25.9 ms ] threadNo comments yet.