# AdaState: Self-Evolving Anchors for Streaming Video Generation arXiv: https://arxiv.org/abs/2605.30349 ยท HTML: https://mlsanity.com/paper/2605.30349/ ## Key facts - Cited by 3 papers in the map (data through 14 Oct 2026). - Placed in Video generation, one of 14 directions in Diffusion models. - Builds on 32 of its 35 references that are in the map. - First posted on arXiv 28 May 2026. ## Placed in - Diffusion models / [Video generation](https://mlsanity.com/diffusion/video-generation/) ## Builds on (most cited first, 30 of 32) - [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023) - [Scaling Rectified Flow Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.03206/) (2024) - [Wan: Open and Advanced Large-Scale Video Generative Models](https://mlsanity.com/paper/2503.20314/) (2025) - [Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling](https://mlsanity.com/paper/1412.3555/) (2014) - [Efficient Streaming Language Models with Attention Sinks](https://mlsanity.com/paper/2309.17453/) (2023) - [One-step Diffusion with Distribution Matching Distillation](https://mlsanity.com/paper/2311.18828/) (2023) - [Show Your Work: Scratchpads for Intermediate Computation with Language Models](https://mlsanity.com/paper/2112.00114/) (2021) - [Training Large Language Models to Reason in a Continuous Latent Space](https://mlsanity.com/paper/2412.06769/) (2024) - [Movie Gen: A Cast of Media Foundation Models](https://mlsanity.com/paper/2410.13720/) (2024) - [Self Forcing: Bridging the Train-Test Gap in Autoregressive Video Diffusion](https://mlsanity.com/paper/2506.08009/) (2025) - [From Slow Bidirectional to Fast Autoregressive Video Diffusion Models](https://mlsanity.com/paper/2412.07772/) (2024) - [Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach](https://mlsanity.com/paper/2502.05171/) (2025) - [Learning to (Learn at Test Time): RNNs with Expressive Hidden States](https://mlsanity.com/paper/2407.04620/) (2024) - [Titans: Learning to Memorize at Test Time](https://mlsanity.com/paper/2501.00663/) (2024) - [LongLive: Real-time Interactive Long Video Generation](https://mlsanity.com/paper/2509.22622/) (2025) - [Think before you speak: Training Language Models With Pause Tokens](https://mlsanity.com/paper/2310.02226/) (2023) - [Rolling Forcing: Autoregressive Long Video Diffusion in Real Time](https://mlsanity.com/paper/2509.25161/) (2025) - [CODI: Compressing Chain-of-Thought into Continuous Space via Self-Distillation](https://mlsanity.com/paper/2502.21074/) (2025) - [Self-Forcing++: Towards Minute-Scale High-Quality Video Generation](https://mlsanity.com/paper/2510.02283/) (2025) - [Let's Think Dot by Dot: Hidden Computation in Transformer Language Models](https://mlsanity.com/paper/2404.15758/) (2024) - [Causal Forcing: Autoregressive Diffusion Distillation Done Right for High-Quality Real-Time Interactive Video Generation](https://mlsanity.com/paper/2602.02214/) (2026) - [VisionReward: Fine-Grained Multi-Dimensional Human Preference Learning for Image and Video Generation](https://mlsanity.com/paper/2412.21059/) (2024) - [Test-Time Training Done Right](https://mlsanity.com/paper/2505.23884/) (2025) - [Infinity-RoPE: Action-Controllable Infinite Video Generation Emerges From Autoregressive Self-Rollout](https://mlsanity.com/paper/2511.20649/) (2025) - [Reward Forcing: Efficient Streaming Video Generation with Rewarded Distribution Matching Distillation](https://mlsanity.com/paper/2512.04678/) (2025) - [Deep Forcing: Training-Free Long Video Generation with Deep Sink and Participative Compression](https://mlsanity.com/paper/2512.05081/) (2025) - [VideoSSM: Autoregressive Long Video Generation with Hybrid State-Space Memory](https://mlsanity.com/paper/2512.04519/) (2025) - [Rolling Sink: Bridging Limited-Horizon Training and Open-Ended Testing in Autoregressive Video Diffusion](https://mlsanity.com/paper/2602.07775/) (2026) - [MemRoPE: Training-Free Infinite Video Generation via Evolving Memory Tokens](https://mlsanity.com/paper/2603.12513/) (2026) - [PackForcing: Short Video Training Suffices for Long Video Sampling and Long Context Inference](https://mlsanity.com/paper/2603.25730/) (2026) ## Built on by (most cited first, 3 of 3) - [Tether the Subject, Release the Scene: Query-Aware Memory Routing for Long-Horizon Autoregressive Video Generation](https://mlsanity.com/paper/2608.26902/) (2026) - [LongTake: Learning to Sustain Dynamics in Long-Horizon Video Generation](https://mlsanity.com/paper/2609.38562/) (2026) - [Custom Forcing: Training-Free Subject Customization for Autoregressive Video Generation](https://mlsanity.com/paper/2610.02914/) (2026) ## Abstract (arXiv) Autoregressive video diffusion models generate streaming video by producing frames sequentially, conditioning each chunk on previously generated content. These models are structurally anchored to the first frame: its key-value representation occupies a privileged position in the attention cache and serves as the primary scene reference throughout generation. As the cleanest and most error-free position in the cache, this anchor draws disproportionate attention, suppressing video dynamics, and locking scene composition to the initial viewpoint even as the scene naturally evolves. The result is a temporally shallow video in which motion, camera movement, and scene progression are dampened in favor of static consistency. To address this, we replace the static anchor with an adaptive state, a hidden latent that the model denoises alongside content at every chunk but never renders. Rather than referencing a frozen first frame, the model generates its own scene anchor at each step by attending to both the previous state and the current content, producing a reference that evolves with the generated content. Unlike standard video generation, which encodes an absolute notion of time, our formulation treats time as relative: every generation step sees the same positional structure regardless of how far generation has progressed, and the state transition is identical at every chunk. Together, these properties introduce a recurrence into the generation process, where denoising serves as the transition function, and the KV cache serves as the carrier, requiring no external module. Experiments demonstrate that the adaptive state substantially improves video dynamics, enabling richer motion and natural scene progression within generated videos.