v-jepa2

VL-JEPA (Vision-Language Joint Embedding Predictive Architecture) in MLX

Reproduction of Interpreting Physics in Video World Models (PEZ, arXiv:2602.07050) — layer-wise linear probing of V-JEPA 2 physics representations

Local-first Notion health tracker with TRIBEv2 brain analysis, AI health insights, symptom logging, goals, medications, appointments, and a browser UI

AI memory for the physical world — 4D spatiotemporal vector database that remembers where things were, not just what's visible now. Hilbert bucketing · temporal sharding · predict-then-retrieve · novelty detection.

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.

A physics-based video search engine using Meta's V-JEPA 2 world model to find videos with similar motion dynamics.

Grounding Spatial Representations via Visual World Models for Robust Autoregressive Drift Reduction in Low-Power Robot Telemetry

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.

Real-time Traffic Anomaly Detection With JEPA