v-jepa

Here are 6 public repositories matching this topic...

mandarwagh9 / dvd-jepa

A tiny, fully-reproducible JEPA world model that learns the physics of a bouncing DVD logo in representation space, dreams its future, and detects anomalies. Trains on a CPU in ~10s. Interactive browser demo. CA: 0x42bef487C250dd054035d5E8d69C12549d1F7Ba3

machine-learning deep-learning reproducible-research pytorch representation-learning anomaly-detection self-supervised-learning video-prediction world-models interactive-demo jepa i-jepa v-jepa

Updated Jun 13, 2026
Python

grisuno / TopoVJEPA

Star

A quaternion-enhanced Video Joint-Embedding Predictive Architecture with continuous spectral autoencoders, Fourier-domain topology, and a 2D torus latent space for world model learning.

topological video-generation quaternion-neural-networks topological-deep-learning world-model v-jepa2 v-jepa

Updated Aug 25, 2026
Python

DaveLeongGithub / jepa-carracing

Star

Frozen V-JEPA 2 (300M params) + PPO policy head achieves +763 mean reward on CarRacing-v3 — 2.4x over CNN baseline, single RTX 3080

reinforcement-learning gymnasium ppo world-models v-jepa video-foundation-models

Updated Mar 18, 2026
Python

GuoCheng24 / world-model-map

Star

A researcher map of open-source world models - what exists, what each line of work claims, and where its authors say it breaks.

research robotics awesome-list model-based-rl self-supervised-learning world-models jepa v-jepa

Updated Aug 25, 2026

mkdsrt10 / mini-v-jepa-from-scratch

Star

Educational implementation of V-JEPA (Video Joint Embedding Predictive Architecture) in PyTorch. Includes Conv3D tubelet embeddings, Video ViT, EMA target encoder, spatiotemporal masking, linear probing, retrieval evaluation, and effective-rank analysis on Something-Something V2.

representation-learning self-supervised-learning world-models video-transformer jepa something-something-v2 ssv2 v-jepa2 v-jepa self-supervised-learning-computer-vision