# A Frame is Worth One Token: Efficient Generative World Modeling with Delta Tokens

arXiv: https://arxiv.org/abs/2604.04913 · HTML: https://mlsanity.com/paper/2604.04913/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Placed in Video-generation world models and simulators, one of 11 directions in World models.
- Builds on 69 of its 87 references that are in the map.
- First posted on arXiv 6 Apr 2026.

## Placed in

- World models / [Video-generation world models and simulators](https://mlsanity.com/wm/video-generation-world-models-and-simulators/)

## Builds on (most cited first, 30 of 69)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [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)
- [Diffusion Models Beat GANs on Image Synthesis](https://mlsanity.com/paper/2105.05233/) (2021)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [The Cityscapes Dataset for Semantic Urban Scene Understanding](https://mlsanity.com/paper/1604.01685/) (2016)
- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [Taming Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2012.09841/) (2020)
- [Sigmoid Loss for Language Image Pre-Training](https://mlsanity.com/paper/2303.15343/) (2023)
- [LLaVA-OneVision: Easy Visual Task Transfer](https://mlsanity.com/paper/2408.03326/) (2024)
- [Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasets](https://mlsanity.com/paper/2311.15127/) (2023)
- [RAFT: Recurrent All-Pairs Field Transforms for Optical Flow](https://mlsanity.com/paper/2003.12039/) (2020)
- [Dream to Control: Learning Behaviors by Latent Imagination](https://mlsanity.com/paper/1912.01603/) (2019)
- [Scaling Autoregressive Models for Content-Rich Text-to-Image Generation](https://mlsanity.com/paper/2206.10789/) (2022)
- [BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learning](https://mlsanity.com/paper/1805.04687/) (2018)
- [Depth Map Prediction from a Single Image using a Multi-Scale Deep Network](https://mlsanity.com/paper/1406.2283/) (2014)
- [CoCa: Contrastive Captioners are Image-Text Foundation Models](https://mlsanity.com/paper/2205.01917/) (2022)
- [SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features](https://mlsanity.com/paper/2502.14786/) (2025)
- [Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction](https://mlsanity.com/paper/2404.02905/) (2024)
- [WorldSimBench: Towards Video Generation Models as World Simulators](https://mlsanity.com/paper/2410.18072/) (2024)

## Built on by (most cited first, 10 of 10)

- [Learning Visual Feature-Based World Models via Residual Latent Action](https://mlsanity.com/paper/2605.07079/) (2026)
- [MemoryVAM: Integrating Memory into Video Action Model for Robot Manipulation](https://mlsanity.com/paper/2606.20679/) (2026)
- [Uncertainty-Aware World Model for Aerial Image-Goal Navigation](https://mlsanity.com/paper/2608.05597/) (2026)
- [StressDream: Steering Video World Models for Robust Policy Evaluation and Improvement](https://mlsanity.com/paper/2606.00267/) (2026)
- [Self-Supervised Learning of Structured Dynamics from Videos](https://mlsanity.com/paper/2607.21576/) (2026)
- [DeltaWAM: Delta World Action Models for Bimanual Manipulation](https://mlsanity.com/paper/2609.28811/) (2026)
- [DeltaWAM: Change-Centric Visual Foresight via Delta Tokens for an Efficient World-Action Model](https://mlsanity.com/paper/2609.33177/) (2026)
- [Token-World: World Modeling in Vision-Language Model Token Space for Robot Manipulation](https://mlsanity.com/paper/2610.00575/) (2026)
- [Latent-Foresight: End-to-End Learning Predictable Representations for Latent World Models](https://mlsanity.com/paper/2610.01942/) (2026)
- [EvoMem-VLA: State-Evolution Memory for Long-Horizon Robot Manipulation](https://mlsanity.com/paper/2610.05418/) (2026)

## Abstract (arXiv)

Anticipating diverse future states is a central challenge in video world modeling. Discriminative world models produce a deterministic prediction that implicitly averages over possible futures, while existing generative world models remain computationally expensive. Recent work demonstrates that predicting the future in the feature space of a vision foundation model (VFM), rather than a latent space optimized for pixel reconstruction, requires significantly fewer world model parameters. However, most such approaches remain discriminative. In this work, we introduce DeltaTok, a tokenizer that encodes the VFM feature difference between consecutive frames into a single continuous "delta" token, and DeltaWorld, a generative world model operating on these tokens to efficiently generate diverse plausible futures. Delta tokens reduce video from a three-dimensional spatio-temporal representation to a one-dimensional temporal sequence, for example yielding a 1,024x token reduction with 512x512 frames. This compact representation enables tractable multi-hypothesis training, where many futures are generated in parallel and only the best is supervised. At inference, this leads to diverse predictions in a single forward pass. Experiments on dense forecasting tasks demonstrate that DeltaWorld forecasts futures that more closely align with real-world outcomes, while having over 35x fewer parameters and using 2,000x fewer FLOPs than existing generative world models. Code and weights: https://deltatok.github.io.