architecture · first seen 4 Oct 2026
joint-embedding predictive architecture
Also: JEPA, I-JEPA, V-JEPA, JEPA world models, masked latent prediction, joint embedding prediction, Video Joint-Embedding Predictive Architecture
A self-supervised architecture that predicts latent representations of masked inputs rather than reconstructing pixels.
Joint-embedding predictive architecture (JEPA) trains an encoder to predict the latent representations of masked or future parts of an input instead of reconstructing raw pixels or signals. The idea dates to Yann LeCun's 2022 proposal and Meta's I-JEPA and V-JEPA models. The current sources are mostly domain transfers rather than new core ideas. They include EEG foundation models, equity-market representations, brain MRI for Alzheimer's, chest X-ray robustness testing, symbolic math functions, point-cloud video and multimodal time series. A cluster of papers also uses JEPA world models for robot planning, alongside a robo_jepa repository from Meta's facebookresearch. Weekly volume is steady rather than accelerating.
JEPA has become a default template for latent-prediction pretraining and world models, so its spread into medical, financial and robotics data is a sign of consolidation rather than a new idea. The more open question is whether latent planning with JEPA world models holds up outside simulation.
Analyst note (established), 17 Nov 2026.
Mentions per week
PapersCodeDiscussion
Rank 61 on 14 Nov 2026 · 74 mentions in 28 days · 73 independent groups · papers → code → discussion · first seen 6 weeks ago · acceleration +0.11
Earliest sources
- CI-JEPA: A Counterfactual Analysis of Latent Representations in Joint-Embedding Predictive Architectures for Self-Supervised LearningPapers · 4 Oct 2026 · Mintu Dutta, Ritesh Vyas, Mohendra Roy *
- SPERA: Spherical Prior EEG Foundation Model with Geometry- and Frequency-Aware Latent PredictionPapers · 4 Oct 2026 · Minsu Kim, Ye-Sung Kim, Hyeseong Jeon, Wooseok Hyung, Joshua Lee, Chang-Hwan Im
- STOCK-JEPA: Prior-Anchored Latent Revision Representation Learning in Equity MarketsPapers · 4 Oct 2026 · Yizhi Luo, Jiahe Yi, Jianhui Zhang, Shuo Sun
- Pythia: Toward Foundation World Models for Multimodal Time SeriesPapers · 4 Oct 2026 · Xilin Dai, Hongzhou Chen, Yifan Hu, Yiding Liu, Zewei Dong, Jiang-Ming Yang
- BeliefGraph-JEPA: Structured Latent World Models for Action-Conditioned Time SeriesPapers · 4 Oct 2026 · Yue Li, Kangqi Ni, Zhen Tan, Tianlong Chen
- T-JEPA: A Temporal Joint-Embedding Predictive Architecture for Learning Better Remote Sensing RepresentationsPapers · 5 Oct 2026 · Bowen Peng, Li Liu, Yongxiang Liu, Weijie Li, Jie Zhou, Zhen Liu
Latest sources
- SADL: On the Demand-Supply Geometry of Representation CollapsePapers · 16 Nov 2026 · Junhan Sun, Yigui Feng, Mingzhe Liu, Haoyan Xu, Lei Ye, Ziqi Han, Hanru Bai, Yan…
- EviBridge: Encoding Predictive Visual Evidence for Proactive InteractionPapers · 16 Nov 2026 · Zhichao Huang, Changheng Cai, Song Ren, Zhichen Zhao, Hongsen Liu, Chaoyang Zhen…
- JEPTERA: A Platform Agnostic JEPA-Assisted Model Predictive Planner for Urban UAV NavigationPapers · 16 Nov 2026 · Gibran Fathoni Belva, Mansur Arief, Ahmad Alfan Alfian Irfan
- PerturbJEPA: Joint-Embedding Predictive Learning of Cellular Perturbation ResponsesPapers · 16 Nov 2026 · Jiangshuan Pang, Youzhe He, Yiting Deng, Wangyang Tang, Huinan Xu, Qi Lu, Shipin…
- VISTA: A Framework for Vision-Informed Semantic Traffic AwarenessPapers · 16 Nov 2026 · Murat Arda Onsu, Poonam Lohan, Burak Kantarci, Aisha Syed, Matthew Andrews, Sean…
- Latent World Modeling for Point-Cloud Video PretrainingPapers · 15 Nov 2026 · Zhuoyue Zhang, Jian Liu, Jihua Zhu, Ajmal Saeed Mian
- JEPA4R: Embodied JEPA World Models with 4D Point PlanningPapers · 15 Nov 2026 · Jiarui Dong, Zhiheng Liang, Yanjun Li, Zeyu Zhang, Heejeong Nam, Haoyu Chen, Hao…
- Func-JEPA: Learning a Sobolev Space of Functions as Latent VectorsPapers · 14 Nov 2026 · Alejandro Moreno Arcas
- WA-JEPA: Goal-Aware Predictive Feedback Action Refinement for a World-Action ModelPapers · 14 Nov 2026 · Tao An, Weihang Liang, Tianqi Wang, Chuqing Cao, Wenming Tan, Kai Zhang, Haibo L…
- After 600 experiments, here's what I discovered about JEPA world modelsDiscussion · 13 Nov 2026 · halcdev · 295 points
- AG-JEPA: Anatomically Guided Joint Embedding Predictive Architecture for Neurodegenerative Representation LearningPapers · 13 Nov 2026 · Azad Singh, Ayantika Das, Amog Rao, Aiman Farooq, Anirban Porya, Ananya Shukla,…
- Cracks in Embeddings: Robustness and Stability of JEPA-Based Chest X-Ray RepresentationsPapers · 13 Nov 2026 · Zahid Mansur, Kalpana Yadav, Deepak Mishra