Robustness of Video-JEPA Representations with Auxiliary Objectives in Out-of-Distribution Evaluation

Joint-Embedding Predictive Architectures (JEPA) are a promising framework for self-supervised video representation learning, yet the behavior of auxiliary objectives in small-scale Video-JEPA training is not well characterized. We report a small-scale empirical study of 18 auxiliary objective variants for Video-JEPA across two pretraining regimes: single-dataset (UCF-101) and mixed-dataset (UCF-101 + Something-Something V2 + ImageNet-100). We evaluate frozen representations on three complementary benchmarks: Diving-48 (fine-grained motion), SomethingSomething V2 (temporal reasoning), and Image

Research goal: How does the incorporation of auxiliary objectives in Video-JEPA models impact the robustness of learned representations when evaluated on out-of-distribution datasets such as Kinetics-700 compared to Kinetics-400?

Autonomous synthesis report generated by SOVEREIGN Research Kernel. Tribunal consensus score: 8.4/10.