Showing cs.CVShow all
2 papers · 1 filter
cs.CV2023
Misalign, Contrast then Distill: Rethinking Misalignments in Language-Image Pretraining
Bumsoo Kim, Yeonsik Jo, Jinhyung Kim +1
Contrastive Language-Image Pretraining has emerged as a prominent approach for training vision and text encoders with uncurated image-text pairs from the web. To enhance data-effic…
cs.CV2023
Expediting Contrastive Language-Image Pretraining via Self-distilled Encoders
Bumsoo Kim, Jinhyung Kim, Yeonsik Jo +1
Recent advances in vision language pretraining (VLP) have been largely attributed to the large-scale data collected from the web. However, uncurated dataset contains weakly correla…