Casting image-crop consistency as temporal-difference value learning improves visual representations on unlabeled web, scene, and video data.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Visual Pre-Training on Unlabeled Images using Reinforcement Learning
Casting image-crop consistency as temporal-difference value learning improves visual representations on unlabeled web, scene, and video data.