Manish Dhakal

It contains all of the datasets and trained ckpts used the in paper, DistillLens (https://arxiv.org/pdf/2602.13567).