Anchor Forcing: Anchor Memory and Tri-Region RoPE for Interactive Streaming Video Diffusion

• URL: http://arxiv.org/abs/2603.13405v1
• Date: Thu, 12 Mar 2026 07:40:06 GMT
• Title: Anchor Forcing: Anchor Memory and Tri-Region RoPE for Interactive Streaming Video Diffusion
• Abstract summary: We propose textbfAnchor Forcing, a cache-centric framework with two designs.<n>First, an anchor-guided re-cache mechanism stores KV states in anchor caches and warm-starts re-cache from these anchors at each prompt switch.<n>Second, a tri-region RoPE with region-specific reference origins reconciles unbounded streaming indices with the pretrained RoPE regime to better retain motion priors.
• Score: 40.885162663758585
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Interactive long video generation requires prompt switching to introduce new subjects or events, while maintaining perceptual fidelity and coherent motion over extended horizons. Recent distilled streaming video diffusion models reuse a rolling KV cache for long-range generation, enabling prompt-switch interaction through re-cache at each switch. However, existing streaming methods still exhibit progressive quality degradation and weakened motion dynamics. We identify two failure modes specific to interactive streaming generation: (i) at each prompt switch, current cache maintenance cannot simultaneously retain KV-based semantic context and recent latent cues, resulting in weak boundary conditioning and reduced perceptual quality; and (ii) during distillation, unbounded time indexing induces a positional distribution shift from the pretrained backbone's bounded RoPE regime, weakening pretrained motion priors and long-horizon motion retention. To address these issues, we propose \textbf{Anchor Forcing}, a cache-centric framework with two designs. First, an anchor-guided re-cache mechanism stores KV states in anchor caches and warm-starts re-cache from these anchors at each prompt switch, reducing post-switch evidence loss and stabilizing perceptual quality. Second, a tri-region RoPE with region-specific reference origins, together with RoPE re-alignment distillation, reconciles unbounded streaming indices with the pretrained RoPE regime to better retain motion priors. Experiments on long videos show that our method improves perceptual quality and motion metrics over prior streaming baselines in interactive settings. Project page: https://github.com/vivoCameraResearch/Anchor-Forcing

Related papers

• Surprise Forcing: What to Remember, When to Skip in Long Video Generation [65.0022589149937]
  Streaming autoregressive diffusion makes minute-scale video synthesis practical, but its bounded context and fixed denoising schedule allocate resources uniformly.<n>We introduce Surprise Forcing, a training-free framework that treats both limitations as online resource-allocation problems.<n> Experiments on VBench, VBench-Long, and VBench-2.0 show that the proposed allocation strategy improves long-horizon consistency and visual quality while retaining real-time streaming throughput.
  arXiv  Detail & Related papers  (2026-07-20T18:37:58Z)
• SoLAR: Error-Resilient Streamable Long-Horizon Free-Viewpoint Video Reconstruction with Anchor Activation and Latent Recalibration [57.159190580279585]
  Motivated by bit allocation theory, we analyze dynamic-anchor-based volumetric video representation within a rate-distortion optimization framework.<n>We propose textbfSoLAR, which is the first error-resilient streamable FVV framework that maintains stable reconstruction quality on long sequences.
  arXiv  Detail & Related papers  (2026-05-08T06:48:59Z)
• Salt: Self-Consistent Distribution Matching with Cache-Aware Training for Fast Video Generation [27.698320788533405]
  Trajectory-style consistency distillation (DMD) can recover sharp, mode-seeking samples, but its local training signals do not explicitly regularize how denoising updates compose across timesteps.<n>We propose Self-Consistent Distribution Matching Distillation (SC-DMD), which explicitly regularizes the endpoint-consistent composition of consecutive denoising updates.
  arXiv  Detail & Related papers  (2026-04-03T15:43:21Z)
• WorldCache: Content-Aware Caching for Accelerated Video World Models [50.7543797435026]
  We present textbfWorldCache, a Perception-Constrained Dynamic CaCache framework.<n>WorldCache improves both when and how to reuse features.<n>On Cosmos-2.5-2B evaluated on PAI-Bench, WorldCache achieves textbf$2.3times$ inference speedup while preserving textbf99.4% of baseline quality.
  arXiv  Detail & Related papers  (2026-03-23T17:59:54Z)
• AvatarForcing: One-Step Streaming Talking Avatars via Local-Future Sliding-Window Denoising [15.787466786514164]
  AvatarForcing is a one-step streaming diffusion framework that denoises a fixed local-future window with heterogeneous noise levels.<n>Experiments on standard benchmarks and a new 400-video long-form benchmark show strong visual quality and lip synchronization at 34 ms/frame.
  arXiv  Detail & Related papers  (2026-03-15T11:42:07Z)
• MemRoPE: Training-Free Infinite Video Generation via Evolving Memory Tokens [34.228121359393775]
  We introduce MemRoPE, a training-free framework with two co-designed components.<n>Online RoPE Indexing caches unrotated keys and applies positional embeddings dynamically at attention time.<n>MemRoPE outperforms existing methods in temporal coherence, visual fidelity, and subject consistency across minute- to hour-scale generation.
  arXiv  Detail & Related papers  (2026-03-12T23:14:16Z)
• LoL: Longer than Longer, Scaling Video Generation to Hour [50.945885467651216]
  This work achieves the first demonstration of real-time, streaming, and infinite-length video generation with little quality decay.<n>As an illustration, we generate continuous videos up to 12 hours in length, which, to our knowledge, is among the longest publicly demonstrated results in streaming video generation.
  arXiv  Detail & Related papers  (2026-01-23T17:21:35Z)
• Deep Forcing: Training-Free Long Video Generation with Deep Sink and Participative Compression [36.99018442740971]
  We find that naively applying StreamingLLM-style attention sinks to video diffusion leads to fidelity degradation and motion stagnation.<n>We introduce Deep Forcing, which consists of two training-free mechanisms that address this without any fine-tuning.<n>Our results demonstrate that training-free KV-cache management can match or exceed training-based approaches for autoregressively streaming long-video generation.
  arXiv  Detail & Related papers  (2025-12-04T18:46:44Z)
• Reward Forcing: Efficient Streaming Video Generation with Rewarded Distribution Matching Distillation [69.57572900337176]
  We introduce Reward Forcing, a novel framework for efficient streaming video generation.<n> EMA-Sink tokens capture both long-term context and recent dynamics, preventing initial frame copying.<n>Re-DMD biases the model's output distribution toward high-reward regions by prioritizing samples with greater dynamics rated by a vision-language model.
  arXiv  Detail & Related papers  (2025-12-04T11:12:13Z)
• Infinity-RoPE: Action-Controllable Infinite Video Generation Emerges From Autoregressive Self-Rollout [15.899488263212442]
  $infty$-RoPE is a unified inference-time framework for autoregressive video diffusion.<n>Block-Relativistic RoPE reformulates temporal encoding as a moving local reference frame.<n> KV Flush renews the KV cache by retaining only two latent frames, the global sink and the last generated latent frame.<n>RoPE Cut introduces controlled discontinuities in temporal RoPE coordinates.
  arXiv  Detail & Related papers  (2025-11-25T18:59:46Z)
• Lightning Fast Caching-based Parallel Denoising Prediction for Accelerating Talking Head Generation [50.04968365065964]
  Diffusion-based talking head models generate high-quality, photorealistic videos but suffer from slow inference.<n>We introduce Lightning-fast Caching-based Parallel denoising prediction (LightningCP)<n>We also propose Decoupled Foreground Attention (DFA) to further accelerate attention computations.
  arXiv  Detail & Related papers  (2025-08-25T02:58:39Z)
• DiTVR: Zero-Shot Diffusion Transformer for Video Restoration [48.97196894658511]
  DiTVR is a zero shot video restoration framework that couples a diffusion transformer with trajectory aware attention and a flow consistent sampler.<n>Our attention mechanism aligns tokens along optical flow trajectories, with particular emphasis on vital layers that exhibit the highest sensitivity to temporal dynamics.<n>The flow guided sampler injects data consistency only into low-frequency bands, preserving high frequency priors while accelerating cache.
  arXiv  Detail & Related papers  (2025-08-11T09:54:45Z)
• Live2Diff: Live Stream Translation via Uni-directional Attention in Video Diffusion Models [64.2445487645478]
  Large Language Models have shown remarkable efficacy in generating streaming data such as text and audio. We present Live2Diff, the first attempt at designing a video diffusion model with uni-directional temporal attention, specifically targeting live streaming video translation.
  arXiv  Detail & Related papers  (2024-07-11T17:34:51Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.