Episode
Enhancing Train-Free Infinite-Frame Generation for Consistent Long Videos
- Podcast
- Daily Paper Cast
- Published
- May 22, 2026
- Duration seconds
- 1501
- Processing state
not_requested- Canonical source
- https://share.transistor.fm/s/a48cc988
Actions
POST https://stenobird.com/v1/public/podcasts/daily-paper-cast-7079649/episodes/enhancing-train-free-infinite-frame-generation-for-consistent-long-videos/transcription-requests
Idempotently request low-priority transcript generation for this episode.GET https://stenobird.com/podcast/daily-paper-cast-7079649/enhancing-train-free-infinite-frame-generation-for-consistent-long-videos.md
Read the agent-friendly Markdown representation of this episode resource.
Summary
đŸ¤— Upvotes: 83 | cs.CV Authors: X. Feng, J. Zhu, M. Wu, C. Chen, F. Mao, H. Guo, J. Wu, X. Chu, K. Huang Title: Enhancing Train-Free Infinite-Frame Generation for Consistent Long Videos Arxiv: http://arxiv.org/abs/2605.18233v1 Abstract: Without incurring significant computational overhead, train-free long video generation aims to enable foundation video generation models to produce longer videos. Frame-level autoregressive frameworks, e.g., FIFO-diffusion, offer the advantage of generating infinitely long videos with constant memory consumption. However, the mismatch between training and inference, coupled with the challenge of maintaining long-term consistency, limits the effective utilization of foundation models. To mitigate these concerns, we propose \textbf{MIGA}, a novel infinite-frame long video generation method. Firstly, we propose an effective two-stage alignment mechanism that mitigates the training-inference gap by reducing the excessive noise span fed to the model. We then introduce an innovative dual consistency enhancement mechanism, where the self-reflection approach corrects early high-noise frames and the long-range frame guidance approach leverages later low-noise frames with broad coverage to steer generation, jointly improving temporal consistency. Extensive experiments on VBench and NarrLV demonstrate the state-of-the-art performance of MIGA. Our project page is available at https://xiaokunfeng.github.io/miga_homepage/.