Stable Video Infinity: Infinite-Length Video Generation with Error Recycling

• URL: http://arxiv.org/abs/2510.09212v1
• Date: Fri, 10 Oct 2025 09:45:46 GMT
• Title: Stable Video Infinity: Infinite-Length Video Generation with Error Recycling
• Authors: Wuyang Li, Wentao Pan, Po-Chien Luan, Yang Gao, Alexandre Alahi,
• Abstract summary: We propose Stable Video Infinity (SVI) that is able to generate infinite-length videos with high temporal consistency, plausible scene transitions, and controllable streaming storylines.<n> SVI incorporates Error-Recycling Fine-Tuning, a new type of efficient training that recycles the Diffusion Transformer's self-generated errors into supervisory prompts.<n>We evaluate SVI on three benchmarks, including consistent, creative, and conditional settings, thoroughly verifying its versatility and state-of-the-art role.
• Score: 76.91310169118408
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We propose Stable Video Infinity (SVI) that is able to generate infinite-length videos with high temporal consistency, plausible scene transitions, and controllable streaming storylines. While existing long-video methods attempt to mitigate accumulated errors via handcrafted anti-drifting (e.g., modified noise scheduler, frame anchoring), they remain limited to single-prompt extrapolation, producing homogeneous scenes with repetitive motions. We identify that the fundamental challenge extends beyond error accumulation to a critical discrepancy between the training assumption (seeing clean data) and the test-time autoregressive reality (conditioning on self-generated, error-prone outputs). To bridge this hypothesis gap, SVI incorporates Error-Recycling Fine-Tuning, a new type of efficient training that recycles the Diffusion Transformer (DiT)'s self-generated errors into supervisory prompts, thereby encouraging DiT to actively identify and correct its own errors. This is achieved by injecting, collecting, and banking errors through closed-loop recycling, autoregressively learning from error-injected feedback. Specifically, we (i) inject historical errors made by DiT to intervene on clean inputs, simulating error-accumulated trajectories in flow matching; (ii) efficiently approximate predictions with one-step bidirectional integration and calculate errors with residuals; (iii) dynamically bank errors into replay memory across discretized timesteps, which are resampled for new input. SVI is able to scale videos from seconds to infinite durations with no additional inference cost, while remaining compatible with diverse conditions (e.g., audio, skeleton, and text streams). We evaluate SVI on three benchmarks, including consistent, creative, and conditional settings, thoroughly verifying its versatility and state-of-the-art role.

Related papers

• TokenTrim: Inference-Time Token Pruning for Autoregressive Long Video Generation [45.36298679288268]
  Auto-regressive video generation enables long video synthesis by iteratively conditioning each new batch of frames on previously generated content.<n>Recent work has shown that such pipelines suffer from severe temporal drift, where errors accumulate and amplify over long horizons.<n>We propose a simple, inference-time method that mitigates temporal drift by identifying and removing unstable latent tokens before they are reused for conditioning.
  arXiv  Detail & Related papers  (2026-01-30T19:44:16Z)
• Efficient Video Sampling: Pruning Temporally Redundant Tokens for Faster VLM Inference [5.146388234814547]
  Long videos often exceed the token budget of modern language models, leading to severe context limitations and latency issues.<n>We introduce Efficient Video Sampling (EVS), a simple, plug-and-play method for reducing token redundancy in videos by identifying and pruning temporally static patches.<n>EVS substantially reduces token count while maintaining semantic fidelity, enabling faster inference and longer input sequences.
  arXiv  Detail & Related papers  (2025-10-16T12:34:38Z)
• 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)
• Learning from Streaming Video with Orthogonal Gradients [62.51504086522027]
  We address the challenge of representation learning from a continuous stream of video as input, in a self-supervised manner.<n>This differs from the standard approaches to video learning where videos are chopped and shuffled during training in order to create a non-redundant batch.<n>We demonstrate the drop in performance when moving from shuffled to sequential learning on three tasks.
  arXiv  Detail & Related papers  (2025-04-02T17:59:57Z)
• Temporal-Consistent Video Restoration with Pre-trained Diffusion Models [51.47188802535954]
  Video restoration (VR) aims to recover high-quality videos from degraded ones.<n>Recent zero-shot VR methods using pre-trained diffusion models (DMs) suffer from approximation errors during reverse diffusion and insufficient temporal consistency.<n>We present a novel a Posterior Maximum (MAP) framework that directly parameterizes video frames in the seed space of DMs, eliminating approximation errors.
  arXiv  Detail & Related papers  (2025-03-19T03:41:56Z)
• Rethinking Video Tokenization: A Conditioned Diffusion-based Approach [58.164354605550194]
  New tokenizer, Diffusion Conditioned-based Gene Tokenizer, replaces GAN-based decoder with conditional diffusion model.<n>We trained using only a basic MSE diffusion loss for reconstruction, along with KL term and LPIPS perceptual loss from scratch.<n>Even a scaled-down version of CDT (3$times inference speedup) still performs comparably with top baselines.
  arXiv  Detail & Related papers  (2025-03-05T17:59:19Z)
• Generative Regression Based Watch Time Prediction for Short-Video Recommendation [36.95095097454143]
  Watch time prediction has emerged as a pivotal task in short video recommendation systems.<n>Recent studies have attempted to address these issues by converting the continuous watch time estimation into an ordinal regression task.<n>We propose a novel Generative Regression (GR) framework that reformulates WTP as a sequence generation task.
  arXiv  Detail & Related papers  (2024-12-28T16:48:55Z)
• Vid-Morp: Video Moment Retrieval Pretraining from Unlabeled Videos in the Wild [30.99346916414286]
  We propose a novel paradigm to reduce annotation costs: pretraining the model on unlabeled, real-world videos.<n>Direct pretraining on imperfect pseudo annotations presents significant challenges, including mismatched sentence-video pairs and imprecise temporal boundaries.<n>We propose the ReCorrect algorithm, which comprises two main phases: semantics-guided refinement and memory-consensus correction.
  arXiv  Detail & Related papers  (2024-12-01T13:49:21Z)
• Transform-Equivariant Consistency Learning for Temporal Sentence Grounding [66.10949751429781]
  We introduce a novel Equivariant Consistency Regulation Learning framework to learn more discriminative representations for each video. Our motivation comes from that the temporal boundary of the query-guided activity should be consistently predicted. In particular, we devise a self-supervised consistency loss module to enhance the completeness and smoothness of the augmented video.
  arXiv  Detail & Related papers  (2023-05-06T19:29:28Z)

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.