# V-JEPA 2.1: Unlocking Dense Features in Video Self-Supervised Learning

arXiv: https://arxiv.org/abs/2603.14482 · HTML: https://mlsanity.com/paper/2603.14482/

## Key facts

- Cited by 117 papers in the map (data through 14 Oct 2026).
- Builds on 113 of its 134 references that are in the map.
- First posted on arXiv 15 Mar 2026.

## Builds on (most cited first, 30 of 113)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Bootstrap your own latent: A new approach to self-supervised Learning](https://mlsanity.com/paper/2006.07733/) (2020)
- [The Cityscapes Dataset for Semantic Urban Scene Understanding](https://mlsanity.com/paper/1604.01685/) (2016)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset](https://mlsanity.com/paper/1705.07750/) (2017)
- [Fast R-CNN](https://mlsanity.com/paper/1504.08083/) (2015)
- [Exploring Simple Siamese Representation Learning](https://mlsanity.com/paper/2011.10566/) (2020)
- [Unsupervised Learning of Visual Features by Contrasting Cluster Assignments](https://mlsanity.com/paper/2006.09882/) (2020)
- [BEiT: BERT Pre-Training of Image Transformers](https://mlsanity.com/paper/2106.08254/) (2021)
- [The Kinetics Human Action Video Dataset](https://mlsanity.com/paper/1705.06950/) (2017)
- [SlowFast Networks for Video Recognition](https://mlsanity.com/paper/1812.03982/) (2018)
- [Vision Transformers for Dense Prediction](https://mlsanity.com/paper/2103.13413/) (2021)
- [Is Space-Time Attention All You Need for Video Understanding?](https://mlsanity.com/paper/2102.05095/) (2021)
- [ViViT: A Video Vision Transformer](https://mlsanity.com/paper/2103.15691/) (2021)
- [An Empirical Study of Training Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.02057/) (2021)
- [Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling](https://mlsanity.com/paper/2412.05271/) (2024)
- [Unsupervised Representation Learning by Predicting Image Rotations](https://mlsanity.com/paper/1803.07728/) (2018)
- [Ego4D: Around the World in 3,000 Hours of Egocentric Video](https://mlsanity.com/paper/2110.07058/) (2021)
- [World Models](https://mlsanity.com/paper/1803.10122/) (2018)
- [Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfer](https://mlsanity.com/paper/1907.01341/) (2019)

## Built on by (most cited first, 30 of 117)

- [Being-H0.7: A Latent World-Action Model from Egocentric Videos](https://mlsanity.com/paper/2605.00078/) (2026)
- [World Action Models: The Next Frontier in Embodied AI](https://mlsanity.com/paper/2605.12090/) (2026)
- [World Model for Robot Learning: A Comprehensive Survey](https://mlsanity.com/paper/2605.00080/) (2026)
- [Reconstruction or Semantics? What Makes a Latent Space Useful for Robotic World Models](https://mlsanity.com/paper/2605.06388/) (2026)
- [How Should World Models Be Evaluated for Embodied Decision-Making? A Decision-Making-Centric Position](https://mlsanity.com/paper/2606.15032/) (2026)
- [Kairos: A Regret-Aware Native World-Action Model Stack for Physical AI](https://mlsanity.com/paper/2606.16533/) (2026)
- [Is Forward Prediction Enough? Physical State Grounding for JEPA World Models](https://mlsanity.com/paper/2608.06799/) (2026)
- [Orca: The World is in Your Mind](https://mlsanity.com/paper/2606.30534/) (2026)
- [BVB: Benchmarking Agentic Video Understanding via Programmatic Reconstruction in Blender](https://mlsanity.com/paper/2609.15478/) (2026)
- [World-Action Models for Robot Learning and Control: A Survey](https://mlsanity.com/paper/2609.16074/) (2026)
- [Nano World Models: A Minimalist Implementation of Future Video Prediction](https://mlsanity.com/paper/2605.23993/) (2026)
- [InternVideo3: Agentify Foundation Models with Multimodal Contextual Reasoning](https://mlsanity.com/paper/2606.12195/) (2026)
- [DA-WAM: Decision-Aligned Future Latents for Driving World Models](https://mlsanity.com/paper/2608.19085/) (2026)
- [GE-Act 2.0: Pretraining and Scaling a World-Action Model for Robotic Manipulation](https://mlsanity.com/paper/2609.05588/) (2026)
- [Learning Visual Feature-Based World Models via Residual Latent Action](https://mlsanity.com/paper/2605.07079/) (2026)
- [Towards Spatial Supersensing in the Wild](https://mlsanity.com/paper/2607.13681/) (2026)
- [Latent State Design for World Models under Sufficiency Constraints](https://mlsanity.com/paper/2605.01694/) (2026)
- [EgoExo-WM: Unlocking Exo Video for Ego World Models](https://mlsanity.com/paper/2605.15477/) (2026)
- [Latent Video Prediction Learns Better World Models](https://mlsanity.com/paper/2605.15618/) (2026)
- [VideoRAE: Taming Video Foundation Models for Generative Modeling via Representation Autoencoders](https://mlsanity.com/paper/2607.14088/) (2026)
- [V-RAE: Rethinking Video Latent Spaces for Generation](https://mlsanity.com/paper/2608.13556/) (2026)
- [LeVJEPA: Efficient & Scalable Video Pretraining without the Heuristics](https://mlsanity.com/paper/2608.27395/) (2026)
- [Lifting Embodied World Models for Planning and Control](https://mlsanity.com/paper/2604.26182/) (2026)
- [VISTA: Technical Report for the Ego4D Short-Term Object Interaction Anticipation at EgoVis 2026](https://mlsanity.com/paper/2605.20901/) (2026)
- [JFAA: Technical Report for the EPIC-KITCHENS-100 Action Anticipation Challenge at EgoVis 2026](https://mlsanity.com/paper/2605.20904/) (2026)
- [Learn from your own latents and not from tokens: A sample-complexity theory](https://mlsanity.com/paper/2605.27734/) (2026)
- [GLINT: Sparsely Gated Vision-Language Alignment for Fine-Grained Radiology Representations](https://mlsanity.com/paper/2606.03180/) (2026)
- [CLEAR: Cognition and Latent Evaluation for Adaptive Routing in End-to-End Autonomous Driving](https://mlsanity.com/paper/2606.06219/) (2026)
- [Off-Manifold Refinement: Guiding Video Generators with a Frozen World Model](https://mlsanity.com/paper/2608.29904/) (2026)
- [LookWhen? Fast Video Recognition by Learning When, Where, and What to Compute](https://mlsanity.com/paper/2605.06809/) (2026)

## Abstract (arXiv)

We present V-JEPA 2.1, a family of self-supervised models that learn dense, high-quality visual representations for both images and videos while retaining strong global scene understanding. The approach combines four key components. First, a dense predictive loss uses a masking-based objective in which both visible and masked tokens contribute to the training signal, encouraging explicit spatial and temporal grounding. Second, deep self-supervision applies the self-supervised objective hierarchically across multiple intermediate encoder layers to improve representation quality. Third, multi-modal tokenizers enable unified training across images and videos. Finally, the model benefits from effective scaling in both model capacity and training data. Together, these design choices produce representations that are spatially structured, semantically coherent, and temporally consistent. Empirically, V-JEPA 2.1 achieves state-of-the-art performance on several challenging benchmarks, including 7.71 mAP on Ego4D for short-term object-interaction anticipation and 40.8 Recall@5 on EPIC-KITCHENS for high-level action anticipation, as well as a 20-point improvement in real-robot grasping success rate over V-JEPA-2 AC. The model also demonstrates strong performance in robotic navigation (5.687 ATE on TartanDrive), depth estimation (0.307 RMSE on NYUv2 with a linear probe), and global recognition (77.7 on Something-Something-V2). These results show that V-JEPA 2.1 significantly advances the state of the art in dense visual understanding and world modeling.