Experiments in Joint Embedding Predictive Architectures (JEPAs).
-
Updated
Jan 5, 2024 - Python
#
i-jepa
Here are 10 public repositories matching this topic...
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
A toolkit for research on multimodal representation learning
clip multi-task-learning multimodal-learning contrastive-learning zero-shot-retrieval zero-shot-classification i-jepa
-
Updated
Sep 7, 2026 - Python
A clean, from-scratch PyTorch implementation of I-JEPA trained on STL-10. Built to benchmark representation-space vs. pixel-space prediction against MAE.
computer-vision deep-learning pytorch vit representation-learning stl10 self-supervised-learning vision-transformer i-jepa
-
Updated
Apr 18, 2026 - Jupyter Notebook
A simple and efficient implementation of Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture (I-JEPA)
computer-vision representation-learning self-supervised-learning self-distillation joint-embedding jepa i-jepa
-
Updated
Aug 16, 2024 - Python
A research-oriented project focusing on the practical application of the I-JEPA for instance level classification on multi-class images.
classification lenet vit representation-learning embedding instance-level multimodal vision-transformer jepa i-jepa
-
Updated
May 4, 2026 - Python
Comparative analysis of monocular depth estimation methods (ResNet-50, frozen Stable Diffusion UNet, I-JEPA, SD+I-JEPA fusion, DepthAnything V2) with robustness evaluation under fog, blur, and low-light on NYU Depth V2.
computer-vision depth-estimation nyu-depth-v2 robustness-analysis stable-diffusion i-jepa depth-anything-v2
-
Updated
Jul 5, 2026 - Jupyter Notebook
JEPA-guided continuation for improving black-box transfer of released dSVA adversarial generators
-
Updated
May 11, 2026 - Jupyter Notebook
Reimplement the I-JEPA paper from scratch with minimum necessary code.
-
Updated
Sep 10, 2026 - Python
Retrieve Steam games with similar store banners, with Meta AI's I-JEPA.
game steam games steam-games discovery steam-game i-jepa fb-ijepa facebook-ijepa ijepa meta-ijepa meta-ai-ijepa fb-i-jepa facebook-i-jepa meta-i-jepa meta-ai-i-jepa
-
Updated
Jul 23, 2023
Add this topic to your repo
To associate your repository with the i-jepa topic, visit your repo's landing page and select "manage topics."