Pith. sign in

REVIEW 3 major objections 5 minor 2 cited by

LiveEdit turns bidirectional video editors into a causal 4-step streaming model that edits at 12.66 FPS while keeping unedited backgrounds stable.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · grok-4.5

2026-07-14 17:14 UTC pith:Q3SFXGZQ

load-bearing objection Clean systems fix for real-time region-preserving streaming video editing: staged bidir-to-causal-to-4-step DMD plus SA-only mask cache hits 12.66 FPS with SOTA streaming quality on their bench. the 3 major comments →

arxiv 2606.26740 v2 pith:Q3SFXGZQ submitted 2026-06-25 cs.CV

LiveEdit: Towards Real-Time Diffusion-Based Streaming Video Editing

Xinyu Wang , Chongbo Zhao , Fangneng Zhan , Yue Ma This is my paper
classification cs.CV
keywords streaming video editingdiffusion distillationcausal DiTmask cachereal-time inferencevideo-to-video editingdistribution matching distillation
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

Practical streaming video editing has been blocked by two problems: attention collapses when future frames disappear, and every frame still pays full compute for static backgrounds. LiveEdit claims both can be solved by a three-stage distillation that moves editing skill from a bidirectional diffusion transformer into a unidirectional 4-step causal editor, then by an AR-oriented mask cache that reuses self-attention features on unedited tokens. On a new 120-pair benchmark the method leads streaming baselines in text alignment and background consistency while running at 12.66 frames per second. A reader who wants interactive or augmented-reality editing cares because the combination finally delivers offline-level fidelity under causal, real-time constraints.

Core claim

A progressive three-stage distillation (foundation tuning, teacher-forcing chunk-wise causal DiT, then 4-step distribution-matching distillation) transfers high-fidelity editing ability into a unidirectional streaming model; combined with an L2-derived mask cache that reuses self-attention features on unedited tokens, the system reaches state-of-the-art streaming quality at 12.66 FPS with strict background preservation.

What carries the argument

Three-stage distillation pipeline plus AR-oriented Mask Cache: the pipeline converts a bidirectional DiT into a 4-step causal editor; the cache extracts an L2 editing mask from the previous chunk and reuses self-attention features for static tokens, pruning roughly 70 percent of redundant computation.

Load-bearing premise

An L2 difference mask taken from the immediately previous chunk remains a reliable router for which tokens can safely reuse cached self-attention features on the current chunk.

What would settle it

Re-run the 120-pair benchmark on clips with large camera or object motion so the previous-chunk mask poorly predicts the current frame; if background consistency or text alignment then falls below the no-cache baseline, the cache claim fails.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

Share X Bluesky LinkedIn Reddit HN

If this is right

  • Interactive and AR applications can apply text-guided local edits with sub-100 ms latency and no background flicker.
  • Existing bidirectional editing models can be converted into causal streaming editors without training a new network from scratch.
  • Self-attention, not FFN layers, is the right place to cache redundant spatial-temporal tokens for streaming editing.
  • A dedicated streaming-editing benchmark becomes the reference for joint quality-and-latency evaluation.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same distillation-plus-selective-cache pattern may transfer to other region-conditioned live tasks such as object removal or style transfer.
  • Highly dynamic camera motion could drop the cache hit rate and erase most of the reported latency gain.
  • Longer untrimmed streams may still accumulate drift that the current short-clip benchmark does not expose.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. LiveEdit proposes a causal, chunk-by-chunk diffusion framework for streaming video editing. It transfers editing capability from a bidirectional DiT (Wan2.1-T2V-1.3B backbone) to a 4-step unidirectional Causal DiT via a three-stage pipeline—Foundation Tuning (MSE on full attention), Teacher Forcing with chunk-wise causal attention, and Distribution Matching Distillation (DMD) that bypasses ODE initialization. An AR-oriented Mask Cache then extracts an L2-based binary mask from the preceding 3-frame chunk, prunes ~70% of tokens, and reuses Self-Attention features for unedited regions. On a private 120-pair benchmark the method reports best or near-best TA/BC/MS/DD/AQ/IQ among streaming baselines, a 20-person user study preference, and 12.66 FPS (79 ms/frame) with the cache.

Significance. Real-time streaming video editing with strict background preservation is a genuine deployment bottleneck for AR and live interactive applications; generation-oriented streaming systems do not transfer cleanly. The progressive three-stage distillation (especially Stage-2 teacher forcing that enables direct DMD initialization without ODE) and the selective SA-only mask cache are concrete, well-motivated engineering contributions. Ablations isolating each stage (Table 2) and cache location (SA vs FFN, Table 3 + Fig. 8 cosine similarities) plus a human ranking study strengthen the empirical case. If the quality–speed claims hold under broader motion and the benchmark is released, the work would be a useful practical advance for interactive video editing.

major comments (3)
  1. Sec. 3.3 and Fig. 5: The AR-oriented Mask Cache is load-bearing for both the quality numbers in Table 1 (Ours W/ Cache) and the 12.66 FPS / 79 ms claim. The mask is lagged (computed from chunk k−1) and thresholded to prune ~70% of tokens; Fig. 5 reports mean Temporal IoU 0.016% and Pixel Diff 0.126%, implying near-stationary edited regions. No motion-stratified mask IoU, quality, or latency breakdown is given (camera pan, subject translation, appearing objects). If the lagged mask mis-routes tokens under non-static backgrounds, either background integrity or the latency claim collapses. A stratified evaluation (or explicit failure cases) is required to support the joint SOTA+real-time claim for interactive/AR use.
  2. Sec. 4.2–4.3 and Table 1: SOTA claims rest on a private 120-pair benchmark with no public release, no description of motion/edit-type diversity, and no error bars or multi-seed statistics. Without these, it is impossible to judge whether the reported margins (e.g., TA 0.270 vs InsV2V 0.259) generalize beyond the low-motion regime suggested by Fig. 5. Releasing the split and reporting variance (or at least motion statistics) is necessary for the comparative claims to be verifiable.
  3. Table 2 / Sec. 4.1: Latency figures (7.89 s for 81 frames without cache; 12.66 FPS with cache) omit resolution, hardware beyond “8 A100s for training,” batching, and whether VAE encode/decode and text encoding are included. Because the real-time suitability claim is central, a precise measurement protocol (resolution, device, end-to-end vs DiT-only) is needed so the 79 ms/frame number can be reproduced and compared fairly to StreamDiffusionV2 et al.
minor comments (5)
  1. Fig. 1 caption and gallery text contain concatenated words (“Changetheskytoagradient…”, “12.05× 97 .38×”); clean for camera-ready.
  2. Sec. 3.2 Stage 3: the DMD gradient formula is written with mixed notation (ϵreal_ϕ vs ϵf ake_ψ); align subscripts and define w(t) explicitly or cite the exact DMD reference used.
  3. Table 2 “Latency 7.89” is labeled in seconds for 81 frames while Fig. 1 reports FPS; state both consistently and note whether CFG is disabled only at Stage 3.
  4. Related Work: EgoEdit and Self-Forcing are discussed in the supplement; a short main-text paragraph clarifying the ODE-bypass and editing-vs-generation distinction would help readers who skip the supplement.
  5. Fig. 7 / Table 3: “Cache on FFN” collapse is clear; add a one-sentence note on whether Cross-Attention was ever cached (the text says only SA is cached).

Circularity Check

0 steps flagged

No circularity: empirical distillation-and-cache pipeline evaluated on held-out pairs against external baselines; no prediction reduces to a fitted input by construction.

full rationale

LiveEdit is an engineering paper that defines a three-stage training procedure (bidirectional foundation MSE, teacher-forced causal DiT, then 4-step DMD) plus an AR-oriented Mask Cache that reuses Self-Attention features for tokens whose L2 difference on the preceding chunk falls below a dynamic threshold. All quantitative claims (Table 1 TA/BC/MS/etc., FPS/latency numbers, ablations in Tables 2-3, user study) are obtained by running the trained model on a 120-pair benchmark and comparing to independently published baselines (InsV2V, LucyEdit, StreamDiffusionV2, etc.). No equation equates a claimed prediction to a parameter that was fitted on the same quantity; the mask is an inference-time routing heuristic whose quality impact is measured, not assumed by definition. Self-citations appear only as ordinary related-work baselines or architectural starting points (Wan2.1, DMD formulations) and are not load-bearing uniqueness theorems. The derivation chain is therefore self-contained and non-circular.

Axiom & Free-Parameter Ledger

4 free parameters · 3 axioms · 2 invented entities

The central performance claims rest on standard diffusion training assumptions plus a handful of engineering choices (chunk size, prune ratio, cache location) that are fitted or selected on the authors’ data. No new physical entities are postulated.

free parameters (4)
  • mask prune ratio / τ = ~70 % prune
    Dynamically set so that ~70 % of spatial tokens are cached; chosen for speed/quality trade-off on the training distribution.
  • chunk size = 3
    Fixed at 3 latent frames for both teacher-forcing and inference; selected for latency vs. temporal context.
  • DMD sampling timesteps = [0,250,500,750]
    Hard-coded to [0,250,500,750] for the 4-step generator.
  • learning rates and step counts = 1e-5; 9k/20k/10k
    Stage-wise AdamW rates (1e-5) and step budgets (9 k / 20 k / 10 k) chosen by the authors.
axioms (3)
  • domain assumption Channel-wise concatenation of noisy and condition latents preserves editing capacity while avoiding quadratic sequence growth.
    Stated in Stage 1; standard practice but not proven optimal for all editing tasks.
  • ad hoc to paper Self-Attention features of unedited regions exhibit high temporal cosine similarity (mean 0.893) and can be safely reused.
    Empirical observation used to justify caching only SA layers (Fig. 8).
  • domain assumption Teacher-forcing with chunk-wise causal masks transfers bidirectional editing priors without future frames.
    Core of Stage 2; relies on the assumption that localized attention can be recovered by distillation.
invented entities (2)
  • AR-oriented Mask Cache independent evidence
    purpose: Dynamically route full computation only to edited tokens and reuse cached SA features for static background.
    New inference mechanism introduced in Sec. 3.3; independent evidence is the ablation showing SA cache preserves quality while FFN cache collapses it.
  • Three-stage distillation pipeline (Foundation → Teacher-Forcing → DMD) independent evidence
    purpose: Progressively convert a bidirectional editor into a 4-step causal streaming editor without ODE initialization.
    Architectural recipe claimed novel relative to Self-Forcing and EgoEdit; evidence is the stage-wise ablation table.

pith-pipeline@v1.1.0-grok45 · 25169 in / 2609 out tokens · 23808 ms · 2026-07-14T17:14:12.200395+00:00 · methodology

0 comments
Cite this review

Pith. "Pith review of LiveEdit: Towards Real-Time Diffusion-Based Streaming Video Editing." pith.science (2026). https://pith.science/paper/Q3SFXGZQ

@misc{pith2026260626740,
  author       = {Pith},
  title        = {Pith review of: LiveEdit: Towards Real-Time Diffusion-Based Streaming Video Editing},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/Q3SFXGZQ}},
  note         = {Machine review of arXiv:2606.26740}
}
Share X Bluesky LinkedIn Reddit HN
read the original abstract

Streaming video editing has made rapid progress, yet practical deployment is still limited by two core issues: maintaining stable backgrounds and non-edited regions over time, and achieving the low latency required for real-time interactive scenarios. Meanwhile, recent streaming video generation methods are mostly developed for synthesis and cannot be directly applied to editing due to the strict preservation requirement and region-specific control. In this work, we present a novel streaming video editing framework that performs causal, frame-by-frame editing with strong content preservation and real-time responsiveness. Our key design is a three-stage distillation pipeline that progressively transfers editing capability from a powerful bidirectional foundation model to an efficient unidirectional streaming editor, enabling stable long-horizon edits without sacrificing visual fidelity. To further support real-time deployment, we introduce an AR-oriented mask cache that reuses region-related computation across frames, substantially reducing redundant processing and accelerating inference. Finally, we establish a dedicated benchmark for streaming video editing. Extensive evaluations demonstrate that our method achieves state-of-the-art visual quality among streaming baselines while drastically boosting inference speed to 12.66 FPS, making it suitable for interactive and augmented reality applications.

Figures

Figures reproduced from arXiv: 2606.26740 by Chongbo Zhao, Fangneng Zhan, Xinyu Wang, Yue Ma.

Figure 1
Figure 1. Figure 1: Gallery of various editing results and efficiency comparisons. We propose the LiveEdit, a novel streaming video editing framework capable of performing causal, chunk-by-chunk manipulation with ultra-low latency and strict background preservation. By synergizing a progressive three-stage architectural distillation pipeline with an AR-oriented Mask Cache, LiveEdit effectively resolves the architectural incom… view at source ↗
Figure 2
Figure 2. Figure 2: Comparison of video editing paradigms. Unlike bidi￾rectional models that suffer from inefficient inference, and past streaming models that fail to preserve accurate unedited content, our proposed streaming editing model leverages a Causal DiT with a mask-guided cache mechanism to achieve high-fidelity and effi￾cient editing. nisms and text embeddings, the model learns complex edit￾ing mappings supervised b… view at source ↗
Figure 3
Figure 3. Figure 3: Visualization of the attention distribution shift. Left: The bidirectional prior exhibits localized attention gather￾ing. Right: Direct causal truncation forces attention to spread uniformly across all historical frames. We summarize the two primary observations regarding the inefficiencies of adapting state-of-the-art video diffusion models to the streaming video editing task and propose the modules to ad… view at source ↗
Figure 4
Figure 4. Figure 4: Overview of the proposed streaming video editing framework. Our approach features a three-stage distillation pipeline that transfers editing capabilities from a bidirectional DiT to a 4-step causal model. Furthermore, an AR-oriented Mask Cache accelerates real-time inference by dynamically decoupling computation and reusing tokens in unedited background regions. causal attention. Let Mcausal denote the cau… view at source ↗
Figure 5
Figure 5. Figure 5: Visualization of the temporal consistency analysis and mask generation process. The left panels show (from top to bottom) the source video frames, the synthesized video frames, the computed difference matrices, and the resulting binary masks. The right panels display the statistical distributions of Temporal IoU and Pixel Difference across the sequence, with mean values of 0.016% and 0.126%, respectively, … view at source ↗
Figure 6
Figure 6. Figure 6: Qualitative comparison of streaming video editing performance. The source videos and instructions are displayed at the top. While existing methods exhibit significant limitations, leading to structural collapse or an inability to accurately follow the text, our approach precisely modifies the target regions and preserves the visual quality and temporal coherence of the original scenes. 4.3. Quantitative co… view at source ↗
Figure 7
Figure 7. Figure 7: Visual comparison of different cache locations. With the instruction ”Change the red currants to deep purple grapes with a thin layer of frost on their skins” [PITH_FULL_IMAGE:figures/full_fig_p009_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Distribution of token cosine similarity between consec [PITH_FULL_IMAGE:figures/full_fig_p010_8.png] view at source ↗
Figure 1
Figure 1. Figure 1: User study results. Volunteers ranked the generated videos from our method and six baselines across three metrics: Instruction Consistency, Background Preservation, and Overall Quality. The line plots indicate the proportion of top-3 selections. Our proposed approach overwhelmingly dominates the evaluations, securing the vast majority of absolute ”Best” rankings across all dimensions.. LucyEdit also exhibi… view at source ↗
Figure 2
Figure 2. Figure 2: More cases generated by LiveEdit [PITH_FULL_IMAGE:figures/full_fig_p017_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: More cases generated by LiveEdit [PITH_FULL_IMAGE:figures/full_fig_p018_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: More cases generated by LiveEdit [PITH_FULL_IMAGE:figures/full_fig_p019_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: More comparison between baseline and our LiveEdit [PITH_FULL_IMAGE:figures/full_fig_p020_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: More comparison between baseline and our LiveEdit [PITH_FULL_IMAGE:figures/full_fig_p021_6.png] view at source ↗

Review history (2 revisions) →

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. JoyAI-Video-Edit: Real-Time Open-Ended Video Editing with Autoregressive Diffusion

    cs.CV 2026-08 conditional novelty 6.0

    A 16B causal autoregressive diffusion system reports real-time 720p video editing at about 30 FPS on one B200 GPU, beating streaming editors and matching several offline systems in benchmark scores.

  2. LiveLight: Real-time Streaming Video Relighting with Interactive Control

    cs.CV 2026-08 conditional novelty 6.0

    A diffusion-based system performs real-time, interactive video relighting by injecting multi-plane light irradiance conditions and streaming latent chunks.

Reference graph

Works this paper leans on

76 extracted references · 26 linked inside Pith · cited by 2 Pith papers

  1. [1]

    Scaling instruction-based video editing with a high- quality synthetic dataset.arXiv preprint arXiv:2510.15742,

    Qingyan Bai, Qiuyu Wang, Hao Ouyang, Yue Yu, Hanlin Wang, Wen Wang, Ka Leong Cheng, Shuailei Ma, Yanhong Zeng, Zichen Liu, Yinghao Xu, Yujun Shen, and Qifeng Chen. Scaling instruction-based video editing with a high- quality synthetic dataset.arXiv preprint arXiv:2510.15742,

    arXiv

  2. [2]

    Training diffusion models with reinforce- ment learning.arXiv preprint arXiv:2305.13301, 2023

    Kevin Black, Michael Janner, Yilun Du, Ilya Kostrikov, and Sergey Levine. Training diffusion models with reinforce- ment learning.arXiv preprint arXiv:2305.13301, 2023. 3

    Pith/arXiv arXiv 2023

  3. [3]

    Stable video diffusion: Scaling latent video diffusion models to large datasets.arXiv preprint arXiv:2311.15127, 2023

    Andreas Blattmann, Tim Dockhorn, Sumith Kulal, Daniel Mendelevitch, Maciej Kilian, Dominik Lorenz, Yam Levi, Zion English, Vikram V oleti, Adam Letts, et al. Stable video diffusion: Scaling latent video diffusion models to large datasets.arXiv preprint arXiv:2311.15127, 2023. 3

    Pith/arXiv arXiv 2023

  4. [4]

    To- ken merging: Your vit but faster.arXiv preprint arXiv:2210.09461, 2022

    Daniel Bolya, Cheng-Yang Fu, Xiaoliang Dai, Peizhao Zhang, Christoph Feichtenhofer, and Judy Hoffman. To- ken merging: Your vit but faster.arXiv preprint arXiv:2210.09461, 2022. 3

    Pith/arXiv arXiv 2022

  5. [5]

    Diffusion forcing: Next-token prediction meets full-sequence diffu- sion.Advances in Neural Information Processing Systems, 37:24081–24125, 2024

    Boyuan Chen, Diego Mart ´ı Mons´o, Yilun Du, Max Sim- chowitz, Russ Tedrake, and Vincent Sitzmann. Diffusion forcing: Next-token prediction meets full-sequence diffu- sion.Advances in Neural Information Processing Systems, 37:24081–24125, 2024. 3

    2024

  6. [6]

    Skyreels-v2: Infinite-length film generative model.arXiv preprint arXiv:2504.13074, 2025

    Guibin Chen, Dixuan Lin, Jiangping Yang, Chunze Lin, Junchen Zhu, Mingyuan Fan, Hao Zhang, Sheng Chen, Zheng Chen, Chengcheng Ma, et al. Skyreels-v2: Infinite-length film generative model.arXiv preprint arXiv:2504.13074, 2025. 3

    Pith/arXiv arXiv 2025

  7. [7]

    Con- textflow: Training-free video object editing via adaptive con- text enrichment.arXiv preprint arXiv:2509.17818, 2025

    Yiyang Chen, Xuanhua He, Xiujun Ma, and Yue Ma. Con- textflow: Training-free video object editing via adaptive con- text enrichment.arXiv preprint arXiv:2509.17818, 2025. 2

    arXiv 2025

  8. [8]

    Dove: Efficient one- step diffusion model for real-world video super-resolution,

    Zheng Chen, Zichen Zou, Kewei Zhang, Xiongfei Su, Xin Yuan, Yong Guo, and Yulun Zhang. Dove: Efficient one- step diffusion model for real-world video super-resolution,

  9. [9]

    Consistent video- to-video transfer using synthetic dataset.arXiv preprint arXiv:2311.00213, 2023

    Jiaxin Cheng, Tianjun Xiao, and Tong He. Consistent video- to-video transfer using synthetic dataset.arXiv preprint arXiv:2311.00213, 2023. 3, 7

    Pith/arXiv arXiv 2023

  10. [10]

    Flashattention: Fast and memory-efficient exact attention with io-awareness.Advances in neural information processing systems, 35:16344–16359, 2022

    Tri Dao, Dan Fu, Stefano Ermon, Atri Rudra, and Christo- pher R ´e. Flashattention: Fast and memory-efficient exact attention with io-awareness.Advances in neural information processing systems, 35:16344–16359, 2022. 3

    2022

  11. [11]

    Compact 3d gaussian splatting for dense visual slam.arXiv preprint arXiv:2403.11247, 2024

    Tianchen Deng, Yaohui Chen, Leyan Zhang, Jianfei Yang, Shenghai Yuan, Jiuming Liu, Danwei Wang, Hesheng Wang, and Weidong Chen. Compact 3d gaussian splatting for dense visual slam.arXiv preprint arXiv:2403.11247, 2024. S1

    Pith/arXiv arXiv 2024

  12. [12]

    Gaussiandwm: 3d gaussian driving world model for unified scene understanding and multi-modal generation

    Tianchen Deng, Xuefeng Chen, Yi Chen, Qu Chen, Yuyao Xu, Lijin Yang, Le Xu, Yu Zhang, Bo Zhang, Wuxiong Huang, and Hesheng Wang. Gaussiandwm: 3d gaussian driving world model for unified scene understanding and multi-modal generation. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 10656–10667, 2026. S1

    2026

  13. [13]

    Tsd-sr: One-step diffusion with target score distillation for real-world image super-resolution

    Linwei Dong, Qingnan Fan, Yihong Guo, Zhonghao Wang, Qi Zhang, Jinwei Chen, Yawei Luo, and Changqing Zou. Tsd-sr: One-step diffusion with target score distillation for real-world image super-resolution. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 23174–23184, 2025. 3

    2025

  14. [14]

    Dit4edit: Dif- fusion transformer for image editing

    Kunyu Feng, Yue Ma, Bingyuan Wang, Chenyang Qi, Haozhe Chen, Qifeng Chen, and Zeyu Wang. Dit4edit: Dif- fusion transformer for image editing. InProceedings of the AAAI Conference on Artificial Intelligence, pages 2969– 2977, 2025. 2

    2025

  15. [15]

    Streamdiffusionv2: A streaming system for dynamic and interactive video generation.arXiv preprint arXiv:2511.07399, 2025

    Tianrui Feng, Zhi Li, Shuo Yang, Haocheng Xi, Muyang Li, Xiuyu Li, Lvmin Zhang, Keting Yang, Kelly Peng, Song Han, et al. Streamdiffusionv2: A streaming system for dynamic and interactive video generation.arXiv preprint arXiv:2511.07399, 2025. 3, 4, 7

    arXiv 2025

  16. [16]

    Pai-studio: Cinematic video background replacement with camera-aware motion

    Heyuan Gao, Bangxun Tang, Yiren Song, Guian Fang, Zijian He, Jie Yang, and Mike Zheng Shou. Pai-studio: Cinematic video background replacement with camera-aware motion. arXiv preprint arXiv:2606.01399, 2026. S1

    Pith/arXiv arXiv 2026

  17. [17]

    Streamingt2v: Con- sistent, dynamic, and extendable long video generation from text

    Roberto Henschel, Levon Khachatryan, Hayk Poghosyan, Daniil Hayrapetyan, Vahram Tadevosyan, Zhangyang Wang, Shant Navasardyan, and Humphrey Shi. Streamingt2v: Con- sistent, dynamic, and extendable long video generation from text. InProceedings of the Computer Vision and Pattern Recognition Conference, pages 2568–2577, 2025. 3

    2025

  18. [18]

    Self forcing: Bridging the train- test gap in autoregressive video diffusion.arXiv preprint arXiv:2506.08009, 2025

    Xun Huang, Zhengqi Li, Guande He, Mingyuan Zhou, and Eli Shechtman. Self forcing: Bridging the train- test gap in autoregressive video diffusion.arXiv preprint arXiv:2506.08009, 2025. 3, S1

    Pith/arXiv arXiv 2025

  19. [19]

    Vbench: Comprehensive bench- mark suite for video generative models

    Ziqi Huang, Yinan He, Jiashuo Yu, Fan Zhang, Chenyang Si, Yuming Jiang, Yuanhan Zhang, Tianxing Wu, Qingyang Jin, Nattapol Chanpaisit, et al. Vbench: Comprehensive bench- mark suite for video generative models. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 21807–21818, 2024. 8

    2024

  20. [20]

    Vace: All-in-one video creation and editing

    Zeyinzi Jiang, Zhen Han, Chaojie Mao, Jingfeng Zhang, Yulin Pan, and Yu Liu. Vace: All-in-one video creation and editing. InProceedings of the IEEE/CVF International Con- ference on Computer Vision, pages 17191–17202, 2025. 2, 3

    2025

  21. [21]

    Editverse: Unifying image and video editing and generation with in-context learning.arXiv preprint arXiv:2509.20360, 2025

    Xuan Ju, Tianyu Wang, Yuqian Zhou, He Zhang, Qing Liu, Nanxuan Zhao, Zhifei Zhang, Yijun Li, Yuanhao Cai, Shaoteng Liu, et al. Editverse: Unifying image and video editing and generation with in-context learning.arXiv preprint arXiv:2509.20360, 2025. 3

    Pith/arXiv arXiv 2025

  22. [22]

    Streamdiffusion: A pipeline-level solution for real-time in- teractive generation

    Akio Kodaira, Chenfeng Xu, Toshiki Hazama, Takanori Yoshimoto, Kohei Ohno, Shogo Mitsuhori, Soichi Sugano, Hanying Cho, Zhijian Liu, Masayoshi Tomizuka, et al. Streamdiffusion: A pipeline-level solution for real-time in- teractive generation. InProceedings of the IEEE/CVF In- ternational Conference on Computer Vision, pages 12371– 12380, 2025. 3, 4, 7

    2025

  23. [23]

    Egoedit: Dataset, real-time streaming model, and benchmark for ego- centric video editing.arXiv preprint arXiv:2512.06065,

    Runjia Li, Moayed Haji-Ali, Ashkan Mirzaei, Chaoyang Wang, Arpit Sahni, Ivan Skorokhodov, Aliaksandr Siarohin, Tomas Jakab, Junlin Han, Sergey Tulyakov, et al. Egoedit: Dataset, real-time streaming model, and benchmark for ego- centric video editing.arXiv preprint arXiv:2512.06065,

    arXiv

  24. [24]

    Vmem: Consistent interactive video scene generation with surfel-indexed view memory

    Runjia Li, Philip Torr, Andrea Vedaldi, and Tomas Jakab. Vmem: Consistent interactive video scene generation with surfel-indexed view memory. InProceedings of the IEEE/CVF International Conference on Computer Vision, pages 25690–25699, 2025. 3

    2025

  25. [25]

    Stable video infinity: Infinite-length video generation with error recycling.arXiv preprint arXiv:2510.09212, 2025

    Wuyang Li, Wentao Pan, Po-Chien Luan, Yang Gao, and Alexandre Alahi. Stable video infinity: Infinite-length video generation with error recycling.arXiv preprint arXiv:2510.09212, 2025. 3

    arXiv 2025

  26. [26]

    Personalive! expressive portrait image anima- tion for live streaming.arXiv preprint arXiv:2512.11253,

    Zhiyuan Li, Chi-Man Pun, Chen Fang, Jue Wang, and Xi- aodong Cun. Personalive! expressive portrait image anima- tion for live streaming.arXiv preprint arXiv:2512.11253,

    arXiv

  27. [27]

    Looking backward: Streaming video-to-video translation with feature banks.arXiv preprint arXiv:2405.15757, 2024

    Feng Liang, Akio Kodaira, Chenfeng Xu, Masayoshi Tomizuka, Kurt Keutzer, and Diana Marculescu. Looking backward: Streaming video-to-video translation with feature banks.arXiv preprint arXiv:2405.15757, 2024. 4, 7

    Pith/arXiv arXiv 2024

  28. [28]

    Cot-edit: Let cot guide instruction video edit- ing

    Sen Liang, Fengbin Guan, Youliang Zhang, Xin Li, and Zhibo Chen. Cot-edit: Let cot guide instruction video edit- ing. InProceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition, pages 37960–37970,

  29. [29]

    Spongebob: Sync-aware harmonious audio-visual generative editing.arXiv preprint arXiv:2605.25193, 2026

    Sen Liang, Cong Wang, Fengbin Guan, Zhentao Yu, Yiting Lu, Yuanzhi Wang, Yuan Zhou, Xin Li, and Zhibo Chen. Spongebob: Sync-aware harmonious audio-visual generative editing.arXiv preprint arXiv:2605.25193, 2026. S1

    Pith/arXiv arXiv 2026

  30. [30]

    Sdxl- lightning: Progressive adversarial diffusion distillation

    Shanchuan Lin, Anran Wang, and Xiao Yang. Sdxl- lightning: Progressive adversarial diffusion distillation. arXiv preprint arXiv:2402.13929, 2024. 3

    Pith/arXiv arXiv 2024

  31. [31]

    Diffusion adversarial post-training for one-step video generation.arXiv preprint arXiv:2501.08316,

    Shanchuan Lin, Xin Xia, Yuxi Ren, Ceyuan Yang, Xuefeng Xiao, and Lu Jiang. Diffusion adversarial post-training for one-step video generation.arXiv preprint arXiv:2501.08316,

    arXiv

  32. [32]

    A survey on cache methods in diffu- sion models: Toward efficient multi-modal generation.arXiv preprint arXiv:2510.19755, 2025

    Jiacheng Liu, Xinyu Wang, Yuqi Lin, Zhikai Wang, Peiru Wang, Peiliang Cai, Qinming Zhou, Zhengan Yan, Zexuan Yan, Zhengyi Shi, et al. A survey on cache methods in diffu- sion models: Toward efficient multi-modal generation.arXiv preprint arXiv:2510.19755, 2025. 3

    arXiv 2025

  33. [33]

    Rolling forcing: Autoregressive long video diffusion in real time.arXiv preprint arXiv:2509.25161, 2025

    Kunhao Liu, Wenbo Hu, Jiale Xu, Ying Shan, and Shijian Lu. Rolling forcing: Autoregressive long video diffusion in real time.arXiv preprint arXiv:2509.25161, 2025. 3

    Pith/arXiv arXiv 2025

  34. [34]

    Instaflow: One step is enough for high-quality diffusion- based text-to-image generation

    Xingchao Liu, Xiwen Zhang, Jianzhu Ma, Jian Peng, et al. Instaflow: One step is enough for high-quality diffusion- based text-to-image generation. InThe Twelfth International Conference on Learning Representations, 2023. 3

    2023

  35. [35]

    Deepcache: Accelerating diffusion models for free

    Xinyin Ma, Gongfan Fang, and Xinchao Wang. Deepcache: Accelerating diffusion models for free. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 15762–15772, 2024. 3

    2024

  36. [36]

    Follow your pose: Pose- guided text-to-video generation using pose-free videos

    Yue Ma, Yingqing He, Xiaodong Cun, Xintao Wang, Siran Chen, Xiu Li, and Qifeng Chen. Follow your pose: Pose- guided text-to-video generation using pose-free videos. In Proceedings of the AAAI Conference on Artificial Intelli- gence, pages 4117–4125, 2024. 3

    2024

  37. [37]

    Follow-your-emoji: Fine-controllable and expressive freestyle portrait animation

    Yue Ma, Hongyu Liu, Hongfa Wang, Heng Pan, Yingqing He, Junkun Yuan, Ailing Zeng, Chengfei Cai, Heung-Yeung Shum, Wei Liu, et al. Follow-your-emoji: Fine-controllable and expressive freestyle portrait animation. InSIGGRAPH Asia 2024 Conference Papers, pages 1–12, 2024. S1

    2024

  38. [38]

    Magic- stick: Controllable video editing via control handle transfor- mations

    Yue Ma, Xiaodong Cun, Sen Liang, Jinbo Xing, Yingqing He, Chenyang Qi, Siran Chen, and Qifeng Chen. Magic- stick: Controllable video editing via control handle transfor- mations. In2025 IEEE/CVF Winter Conference on Applica- tions of Computer Vision (WACV), pages 9385–9395. IEEE,

  39. [39]

    Controllable video generation: A survey.arXiv preprint arXiv:2507.16869,

    Yue Ma, Kunyu Feng, Zhongyuan Hu, Xinyu Wang, Yucheng Wang, Mingzhe Zheng, Xuanhua He, Chenyang Zhu, Hongyu Liu, Yingqing He, et al. Controllable video generation: A survey.arXiv preprint arXiv:2507.16869,

    arXiv

  40. [40]

    Follow-your-creation: Empowering 4d creation through video inpainting.arXiv preprint arXiv:2506.04590, 2025

    Yue Ma, Kunyu Feng, Xinhua Zhang, Hongyu Liu, David Junhao Zhang, Jinbo Xing, Yinhan Zhang, Ayden Yang, Zeyu Wang, and Qifeng Chen. Follow-your-creation: Empowering 4d creation through video inpainting.arXiv preprint arXiv:2506.04590, 2025. 3

    Pith/arXiv arXiv 2025

  41. [41]

    Follow-your-motion: Video motion transfer via efficient spatial-temporal decoupled finetuning.arXiv preprint arXiv:2506.05207, 2025

    Yue Ma, Yulong Liu, Qiyuan Zhu, Ayden Yang, Kunyu Feng, Xinhua Zhang, Zhifeng Li, Sirui Han, Chenyang Qi, and Qifeng Chen. Follow-your-motion: Video motion transfer via efficient spatial-temporal decoupled finetuning.arXiv preprint arXiv:2506.05207, 2025. 3

    arXiv 2025

  42. [42]

    Follow-your-emoji-faster: To- wards efficient, fine-controllable, and expressive freestyle portrait animation.arXiv preprint arXiv:2509.16630, 2025

    Yue Ma, Zexuan Yan, Hongyu Liu, Hongfa Wang, Heng Pan, Yingqing He, Junkun Yuan, Ailing Zeng, Chengfei Cai, Heung-Yeung Shum, et al. Follow-your-emoji-faster: To- wards efficient, fine-controllable, and expressive freestyle portrait animation.arXiv preprint arXiv:2509.16630, 2025. S1

    arXiv 2025

  43. [43]

    Group editing: Edit multiple im- ages in one go.arXiv preprint arXiv:2603.22883, 2026

    Yue Ma, Xinyu Wang, Qianli Ma, Qinghe Wang, Mingzhe Zheng, Xiangpeng Yang, Hao Li, Chongbo Zhao, Jixuan Ying, Harry Yang, et al. Group editing: Edit multiple im- ages in one go.arXiv preprint arXiv:2603.22883, 2026. 3

    arXiv 2026

  44. [44]

    Fastvmt: Eliminat- ing redundancy in video motion transfer.arXiv preprint arXiv:2602.05551, 2026

    Yue Ma, Zhikai Wang, Tianhao Ren, Mingzhe Zheng, Hongyu Liu, Jiayi Guo, Mark Fong, Yuxuan Xue, Zixi- ang Zhao, Konrad Schindler, et al. Fastvmt: Eliminat- ing redundancy in video motion transfer.arXiv preprint arXiv:2602.05551, 2026. 3

    arXiv 2026

  45. [45]

    Null-text inversion for editing real im- ages using guided diffusion models

    Ron Mokady, Amir Hertz, Kfir Aberman, Yael Pritch, and Daniel Cohen-Or. Null-text inversion for editing real im- ages using guided diffusion models. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 6038–6047, 2023. 3

    2023

  46. [46]

    Fatezero: Fus- ing attentions for zero-shot text-based video editing

    Chenyang Qi, Xiaodong Cun, Yong Zhang, Chenyang Lei, Xintao Wang, Ying Shan, and Qifeng Chen. Fatezero: Fus- ing attentions for zero-shot text-based video editing. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 15932–15942, 2023. 2

    2023

  47. [47]

    Learning transferable visual models from natural language supervi- sion

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, et al. Learning transferable visual models from natural language supervi- sion. InInternational conference on machine learning, pages 8748–8763. PmLR, 2021. 8

    2021

  48. [48]

    Progressive distillation for fast sampling of diffusion models.arXiv preprint arXiv:2202.00512, 2022

    Tim Salimans and Jonathan Ho. Progressive distillation for fast sampling of diffusion models.arXiv preprint arXiv:2202.00512, 2022. 3

    Pith/arXiv arXiv 2022

  49. [49]

    Laion-aesthetics, 2022

    Christoph Schuhmann. Laion-aesthetics, 2022. 8

    2022

  50. [50]

    Make-a-video: Text-to-video generation without text-video data.arXiv preprint arXiv:2209.14792,

    Uriel Singer, Adam Polyak, Thomas Hayes, Xi Yin, Jie An, Songyang Zhang, Qiyuan Hu, Harry Yang, Oron Ashual, Oran Gafni, et al. Make-a-video: Text-to-video generation without text-video data.arXiv preprint arXiv:2209.14792,

    Pith/arXiv arXiv

  51. [51]

    Consistency models, 2023

    Yang Song, Prafulla Dhariwal, Mark Chen, and Ilya Sutskever. Consistency models, 2023. 3

    2023

  52. [52]

    Pro- cesspainter: Learning to draw from sequence data

    Yiren Song, Shijie Huang, Chen Yao, Hai Ci, Xiaojun Ye, Jiaming Liu, Yuxuan Zhang, and Mike Zheng Shou. Pro- cesspainter: Learning to draw from sequence data. InSIG- GRAPH Asia 2024 Conference Papers, pages 1–10, 2024. S1

    2024

  53. [53]

    Streamingeffect: Real-time human-centric video effect gen- eration.arXiv preprint arXiv:2605.17019, 2026

    Yiren Song, Cheng Liu, Yuxin Jiang, and Mike Zheng Shou. Streamingeffect: Real-time human-centric video effect gen- eration.arXiv preprint arXiv:2605.17019, 2026

    Pith/arXiv arXiv 2026

  54. [54]

    Vista: Triplet-supervised video style transfer with dif- fusion transformers.arXiv preprint arXiv:2605.17312, 2026

    Yiren Song, Wangzi Yao, Haofan Wang, and Mike Zheng Shou. Vista: Triplet-supervised video style transfer with dif- fusion transformers.arXiv preprint arXiv:2605.17312, 2026. S1

    Pith/arXiv arXiv 2026

  55. [55]

    Lucy edit: Open-weight text-guided video editing, 2025

    DecartAI Team. Lucy edit: Open-weight text-guided video editing, 2025. 2, 3, 7

    2025

  56. [56]

    Magi-1: Autoregressive video genera- tion at scale.arXiv preprint arXiv:2505.13211, 2025

    Hansi Teng, Hongyu Jia, Lei Sun, Lingzhi Li, Maolin Li, Mingqiu Tang, Shuai Han, Tianning Zhang, WQ Zhang, Weifeng Luo, et al. Magi-1: Autoregressive video genera- tion at scale.arXiv preprint arXiv:2505.13211, 2025. 3

    Pith/arXiv arXiv 2025

  57. [57]

    Wan: Open and advanced large-scale video generative models, 2025

    Team Wan, Ang Wang, Baole Ai, Bin Wen, Chaojie Mao, Chen-Wei Xie, Di Chen, Feiwu Yu, Haiming Zhao, Jianx- iao Yang, Jianyuan Zeng, Jiayu Wang, Jingfeng Zhang, Jin- gren Zhou, Jinkai Wang, Jixuan Chen, Kai Zhu, Kang Zhao, Keyu Yan, Lianghua Huang, Mengyang Feng, Ningyi Zhang, Pandeng Li, Pingyu Wu, Ruihang Chu, Ruili Feng, Shiwei Zhang, Siyang Sun, Tao Fan...

    2025

  58. [58]

    Cove: Unleashing the diffusion fea- ture correspondence for consistent video editing.Advances in Neural Information Processing Systems, 37:96541–96565,

    Jiangshan Wang, Yue Ma, Jiayi Guo, Yicheng Xiao, Gao Huang, and Xiu Li. Cove: Unleashing the diffusion fea- ture correspondence for consistent video editing.Advances in Neural Information Processing Systems, 37:96541–96565,

  59. [59]

    Tam- ing rectified flow for inversion and editing.arXiv preprint arXiv:2411.04746, 2024

    Jiangshan Wang, Junfu Pu, Zhongang Qi, Jiayi Guo, Yue Ma, Nisha Huang, Yuxin Chen, Xiu Li, and Ying Shan. Tam- ing rectified flow for inversion and editing.arXiv preprint arXiv:2411.04746, 2024. 2

    Pith/arXiv arXiv 2024

  60. [60]

    Seedvr2: One-step video restora- tion via diffusion adversarial post-training.arXiv preprint arXiv:2506.05301, 2025

    Jianyi Wang, Shanchuan Lin, Zhijie Lin, Yuxi Ren, Meng Wei, Zongsheng Yue, Shangchen Zhou, Hao Chen, Yang Zhao, Ceyuan Yang, et al. Seedvr2: One-step video restora- tion via diffusion adversarial post-training.arXiv preprint arXiv:2506.05301, 2025. 3

    arXiv 2025

  61. [61]

    Videolcm: Video latent consistency model.arXiv preprint arXiv:2312.09109,

    Xiang Wang, Shiwei Zhang, Han Zhang, Yu Liu, Yingya Zhang, Changxin Gao, and Nong Sang. Videolcm: Video latent consistency model.arXiv preprint arXiv:2312.09109,

    Pith/arXiv arXiv

  62. [62]

    Tune-a-video: One-shot tuning of image diffusion models for text-to-video generation

    Jay Zhangjie Wu, Yixiao Ge, Xintao Wang, Stan Weixian Lei, Yuchao Gu, Yufei Shi, Wynne Hsu, Ying Shan, Xiaohu Qie, and Mike Zheng Shou. Tune-a-video: One-shot tuning of image diffusion models for text-to-video generation. In Proceedings of the IEEE/CVF international conference on computer vision, pages 7623–7633, 2023. 2

    2023

  63. [63]

    Freeswim: Revisiting sliding-window atten- tion mechanisms for training-free ultra-high-resolution video generation.arXiv preprint arXiv:2511.14712, 2025

    Yunfeng Wu, Jiayi Song, Zhenxiong Tan, Zihao He, and Songhua Liu. Freeswim: Revisiting sliding-window atten- tion mechanisms for training-free ultra-high-resolution video generation.arXiv preprint arXiv:2511.14712, 2025. S1

    arXiv 2025

  64. [64]

    Vibe: Ultra-high-resolution video synthesis born from pure images.arXiv preprint arXiv:2603.23326, 2026

    Yunfeng Wu, Hongying Cheng, Zihao He, and Songhua Liu. Vibe: Ultra-high-resolution video synthesis born from pure images.arXiv preprint arXiv:2603.23326, 2026

    arXiv 2026

  65. [65]

    Smrabooth: Subject and motion representation alignment for customized video generation

    Xuancheng Xu, Yaning Li, Sisi You, and Bing-Kun Bao. Smrabooth: Subject and motion representation alignment for customized video generation. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 16130–16141, 2026. S1

    2026

  66. [66]

    Sparse videogen2: Accelerate video generation with sparse attention via semantic-aware permutation, 2025

    Shuo Yang, Haocheng Xi, Yilong Zhao, Muyang Li, Jintao Zhang, Han Cai, Yujun Lin, Xiuyu Li, Chenfeng Xu, Kelly Peng, Jianfei Chen, Song Han, Kurt Keutzer, and Ion Stoica. Sparse videogen2: Accelerate video generation with sparse attention via semantic-aware permutation, 2025. 3

    2025

  67. [67]

    Unified video editing with temporal reasoner

    Xiangpeng Yang, Ji Xie, Yiyuan Yang, Yan Huang, Min Xu, and Qiang Wu. Unified video editing with temporal reasoner. arXiv preprint arXiv:2512.07469, 2025. 2, 7

    Pith/arXiv arXiv 2025

  68. [68]

    Cogvideox: Text-to-video diffusion models with an expert transformer.arXiv preprint arXiv:2408.06072, 2024

    Zhuoyi Yang, Jiayan Teng, Wendi Zheng, Ming Ding, Shiyu Huang, Jiazheng Xu, Yuanming Yang, Wenyi Hong, Xiao- han Zhang, Guanyu Feng, et al. Cogvideox: Text-to-video diffusion models with an expert transformer.arXiv preprint arXiv:2408.06072, 2024. 3

    Pith/arXiv arXiv 2024

  69. [69]

    Unic: Unified in-context video editing.arXiv preprint arXiv:2506.04216, 2025

    Zixuan Ye, Xuanhua He, Quande Liu, Qiulin Wang, Xintao Wang, Pengfei Wan, Di Zhang, Kun Gai, Qifeng Chen, and Wenhan Luo. Unic: Unified in-context video editing.arXiv preprint arXiv:2506.04216, 2025. 3

    Pith/arXiv arXiv 2025

  70. [70]

    Im- proved distribution matching distillation for fast image syn- thesis.Advances in neural information processing systems, 37:47455–47487, 2024

    Tianwei Yin, Micha ¨el Gharbi, Taesung Park, Richard Zhang, Eli Shechtman, Fredo Durand, and Bill Freeman. Im- proved distribution matching distillation for fast image syn- thesis.Advances in neural information processing systems, 37:47455–47487, 2024. 3

    2024

  71. [71]

    From slow bidirectional to fast autoregressive video diffusion mod- els

    Tianwei Yin, Qiang Zhang, Richard Zhang, William T Free- man, Fredo Durand, Eli Shechtman, and Xun Huang. From slow bidirectional to fast autoregressive video diffusion mod- els. InProceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition, pages 22963–22974,

  72. [72]

    Controllable text-to-image generation with gpt- 4.arXiv preprint arXiv:2305.18583, 2023

    Tianjun Zhang, Yi Zhang, Vibhav Vineet, Neel Joshi, and Xin Wang. Controllable text-to-image generation with gpt- 4.arXiv preprint arXiv:2305.18583, 2023. 2

    Pith/arXiv arXiv 2023

  73. [73]

    Forecast then calibrate: Feature caching as ode for efficient diffusion transformers

    Shikang Zheng, Liang Feng, Xinyu Wang, Qinming Zhou, Peiliang Cai, Chang Zou, Jiacheng Liu, Yuqi Lin, Junjie Chen, Yue Ma, et al. Forecast then calibrate: Feature caching as ode for efficient diffusion transformers. InProceedings of the AAAI Conference on Artificial Intelligence, pages 13449– 13457, 2026. 3

    2026

  74. [74]

    Compute only 16 tokens in one timestep: Accelerating diffusion transformers with cluster-driven fea- ture caching

    Zhixin Zheng, Xinyu Wang, Chang Zou, Shaobo Wang, and Linfeng Zhang. Compute only 16 tokens in one timestep: Accelerating diffusion transformers with cluster-driven fea- ture caching. InProceedings of the 33rd ACM International Conference on Multimedia, pages 10181–10189, 2025. 3

    2025

  75. [75]

    Causal forcing: Autoregressive diffu- sion distillation done right for high-quality real-time inter- active video generation.arXiv preprint arXiv:2602.02214,

    Hongzhou Zhu, Min Zhao, Guande He, Hang Su, Chongx- uan Li, and Jun Zhu. Causal forcing: Autoregressive diffu- sion distillation done right for high-quality real-time inter- active video generation.arXiv preprint arXiv:2602.02214,

    Pith/arXiv arXiv

  76. [76]

    Flashvsr: Towards real- time diffusion-based streaming video super-resolution.arXiv preprint arXiv:2510.12747, 2025

    Junhao Zhuang, Shi Guo, Xin Cai, Xiaohui Li, Yihao Liu, Chun Yuan, and Tianfan Xue. Flashvsr: Towards real- time diffusion-based streaming video super-resolution.arXiv preprint arXiv:2510.12747, 2025. 3 LiveEditUserinput Change the blue sky to a gradient of deep indigo and magenta with faint constellations. LiveEditUserinput Change the brown horse to a da...

    arXiv 2025