FlowLong: Inference-time Long Video Generation via Manifold-constrained Tweedie Matching

• URL: http://arxiv.org/abs/2605.20910v1
• Date: Wed, 20 May 2026 08:55:37 GMT
• Title: FlowLong: Inference-time Long Video Generation via Manifold-constrained Tweedie Matching
• Abstract summary: We propose a novel but simple inference-time approach for long video generation that is architecture-agnostic and requires no additional training.<n>Our method generates long videos via overlapping sliding windows, where predicted clean samples from adjacent windows are blended via emphTweedie matching to enforce both textbfmanifold constraint and temporal consistency across overlap regions.<n>Our method generates videos several times longer than the native window length while outperforming both training-free and autoregressive baselines in temporal consistency and visual quality, and further extends to audio-video joint generation and text-to-3DGS without any
• Score: 68.01498128172214
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Extending the generation horizon of video diffusion models to long sequences remains a long-standing and important challenge. Existing training-free approaches fall into two categories: extensions of bidirectional models, which are tightly coupled to specific architectures and suffer from quality degradation over long horizons, and autoregressive models, which accumulate drift errors due to exposure bias and tend to produce repetitive motion patterns. To address these issues, we propose a novel but simple inference-time approach for long video generation that is architecture-agnostic and requires no additional training. Our method generates long videos via overlapping sliding windows, where predicted clean samples from adjacent windows are blended via \emph{Tweedie matching} to enforce both \textbf{manifold constraint and temporal consistency} across overlap regions. \emph{Stochastic early-phase sampling} then synchronizes per-window trajectories by injecting fresh noise after each Tweedie matching correction in the high-noise phase, before transitioning to deterministic ODE sampling to preserve fine-grained visual fidelity. Applied to various video generation models, our method generates videos several times longer than the native window length while outperforming both training-free and autoregressive baselines in temporal consistency and visual quality, and further extends to audio-video joint generation and text-to-3DGS without any fine-tuning.

Related papers

• Diff-VF: Training-free High-quality Long Video Generation via Diffusion Model [33.214927542512555]
  Diff-VF is a framework that converts short-video diffusion backbones into long-video generators without modifying or fine-tuning the base model.<n>VBench-Long evaluation results show that Diff-VF achieves a more favorable balance between temporal coherence and motion diversity than base models.
  arXiv  Detail & Related papers  (2026-08-06T12:53:46Z)
• LongE2V: Long-Horizon Event-based Video Reconstruction, Prediction, and Frame Interpolation with Video Diffusion Models [14.33633321388011]
  LongE2V is a novel approach that leverages pre-trained video diffusion priors to handle event-based video reconstruction.<n>By fine-tuning agressive video models, our approach mitigates high data efficiency and superior perceptual quality.<n>Experiments on real-world benchmarks demonstrate that LongE2V outperforms state-of-the-art methods across all three tasks.
  arXiv  Detail & Related papers  (2026-07-09T17:59:52Z)
• MV-Forcing: Long Multi-View Video Generation via 4D-Grounded Spatio-Temporal Self-Forcing [25.612425628475705]
  We present MV-Forcing, a framework that composes temporal and view-wise autoregression within a single diffusion model.<n>We show that MV-Forcing produces geometrically consistent multi-view videos of dynamic scenes at arbitrary lengths.
  arXiv  Detail & Related papers  (2026-07-06T17:54:16Z)
• Streaming Autoregressive Video Generation via Diagonal Distillation [50.13573884115673]
  Autoregressive models offer a natural framework for sequential frame synthesis but require heavy computation to achieve high fidelity.<n>We propose Diagonal Distillation to exploit temporal information across both video chunks and denoising steps.<n>Our method generates a 5-second video in 2.61 seconds (up to 31 FPS), achieving a 277.3x speedup over the undistilled model.
  arXiv  Detail & Related papers  (2026-03-10T10:45:24Z)
• Uniform Discrete Diffusion with Metric Path for Video Generation [103.86033350602908]
  Continuous-space video generation has advanced rapidly, while discrete approaches lag behind due to error accumulation and long-duration inconsistency.<n>We present Uniform generative modeling and present Uniform pAth (URSA), a powerful framework that bridges the gap with continuous approaches for scalable video generation.<n>URSA consistently outperforms existing discrete methods and achieves performance comparable to state-of-the-art continuous diffusion methods.
  arXiv  Detail & Related papers  (2025-10-28T17:59:57Z)
• Rolling Forcing: Autoregressive Long Video Diffusion in Real Time [86.40480237741609]
  Rolling Forcing is a novel video generation technique that enables streaming long videos with minimal error accumulation.<n>Rolling Forcing comes with three novel designs. First, instead of iteratively sampling individual frames, which accelerates error propagation, we design a joint denoising scheme.<n>Second, we introduce the attention sink mechanism into the long-horizon stream video generation task, which allows the model to keep key value states of initial frames as a global context anchor.<n>Third, we design an efficient training algorithm that enables few-step distillation over largely extended denoising windows.
  arXiv  Detail & Related papers  (2025-09-29T17:57:14Z)
• FreeLong++: Training-Free Long Video Generation via Multi-band SpectralFusion [24.48220892418698]
  FreeLong is a training-free framework designed to balance the frequency distribution of long video features during the denoising process.<n>FreeLong achieves this by blending global low-frequency features, which capture holistic semantics across the full video, with local high-frequency features extracted from short temporal windows.<n>FreeLong++ extends FreeLong into a multi-branch architecture with multiple attention branches, each operating at a distinct temporal scale.
  arXiv  Detail & Related papers  (2025-06-30T18:11:21Z)
• STAGE: A Stream-Centric Generative World Model for Long-Horizon Driving-Scene Simulation [42.73124501421074]
  STAGE is an auto-regressive framework that pioneers hierarchical feature coordination and multiphase optimization for sustainable video synthesis.<n>HTFT enhances temporal consistency between video frames throughout the video generation process.<n>We generated 600 frames of high-quality driving videos on the Nuscenes dataset, which far exceeds the maximum length achievable by existing methods.
  arXiv  Detail & Related papers  (2025-06-16T06:53:05Z)
• Tuning-Free Multi-Event Long Video Generation via Synchronized Coupled Sampling [81.37449968164692]
  We propose Synchronized Coupled Sampling (SynCoS), a novel inference framework that synchronizes denoising paths across the entire video.<n>Our approach combines two complementary sampling strategies, which ensure seamless local transitions and enforce global coherence.<n>Extensive experiments show that SynCoS significantly improves multi-event long video generation, achieving smoother transitions and superior long-range coherence.
  arXiv  Detail & Related papers  (2025-03-11T16:43:45Z)
• Generating Long Videos of Dynamic Scenes [66.56925105992472]
  We present a video generation model that reproduces object motion, changes in camera viewpoint, and new content that arises over time. A common failure case is for content to never change due to over-reliance on inductive biases to provide temporal consistency.
  arXiv  Detail & Related papers  (2022-06-07T16:29:51Z)

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.