all AI news
MAFA: Managing False Negatives for Vision-Language Pre-training
June 14, 2024, 4:48 a.m. | Jaeseok Byun, Dohoon Kim, Taesup Moon
cs.CV updates on arXiv.org arxiv.org
Abstract: We consider a critical issue of false negatives in Vision-Language Pre-training (VLP), a challenge that arises from the inherent many-to-many correspondence of image-text pairs in large-scale web-crawled datasets. The presence of false negatives can impede achieving optimal performance and even lead to a significant performance drop. To address this challenge, we propose MAFA (MAnaging FAlse negatives), which consists of two pivotal components building upon the recently developed GRouped mIni-baTch sampling (GRIT) strategy: 1) an efficient …
arxiv cs.ai cs.cv false language pre-training replace training type vision vision-language
More from arxiv.org / cs.CV updates on arXiv.org
Jobs in AI, ML, Big Data
Senior Data Engineer
@ Displate | Warsaw
Senior Principal Software Engineer
@ Oracle | Columbia, MD, United States
Software Engineer for Manta Systems
@ PXGEO | Linköping, Östergötland County, Sweden
DevOps Engineer
@ Teradyne | Odense, DK
LIDAR System Engineer Trainee
@ Valeo | PRAGUE - PRA2
Business Applications Administrator
@ Allegro | Poznań, Poland