This study introduces Predictive Representation Learning (PRL) as a category in self-supervised learning centered on latent prediction of unobserved data, positions JEPA as an example, and reports comparative results for BYOL, MAE, and I-JEPA on similarity and robustness metrics.
A-JEPA: Joint-Embedding Predictive Architecture Can Listen,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
From Alignment to Prediction: A Study of Self-Supervised Learning and Predictive Representation Learning
This study introduces Predictive Representation Learning (PRL) as a category in self-supervised learning centered on latent prediction of unobserved data, positions JEPA as an example, and reports comparative results for BYOL, MAE, and I-JEPA on similarity and robustness metrics.