Pith. sign in

REVIEW 10 cited by

RIFLEx: A Free Lunch for Length Extrapolation in Video Diffusion Transformers

Not yet reviewed by Pith; the record is open.

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

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2502.15894 v3 pith:FUQ7OMNJ submitted 2025-02-21 cs.CV

Min Zhao , Guande He , Yixiao Chen , Hongzhou Zhu , Chongxuan Li , Jun Zhu This is my paper · ORCID
classification cs.CV
keywords extrapolationfrequencyriflexvideovideosdiffusionfreehigh-quality
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Recent advancements in video generation have enabled models to synthesize high-quality, minute-long videos. However, generating even longer videos with temporal coherence remains a major challenge and existing length extrapolation methods lead to temporal repetition or motion deceleration. In this work, we systematically analyze the role of frequency components in positional embeddings and identify an intrinsic frequency that primarily governs extrapolation behavior. Based on this insight, we propose RIFLEx, a minimal yet effective approach that reduces the intrinsic frequency to suppress repetition while preserving motion consistency, without requiring any additional modifications. RIFLEx offers a true free lunch--achieving high-quality 2x extrapolation on state-of-the-art video diffusion transformers in a completely training-free manner. Moreover, it enhances quality and enables 3x extrapolation by minimal fine-tuning without long videos. Project page and codes: https://riflex-video.github.io/.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 10 Pith papers

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

  1. Causal Forcing: Autoregressive Diffusion Distillation Done Right for High-Quality Real-Time Interactive Video Generation

    cs.CV 2026-02 conditional novelty 7.0 of 10

    Causal Forcing uses an autoregressive teacher for ODE initialization in diffusion distillation to close the causal attention gap and deliver better real-time video generation than Self Forcing.

  2. Phase-Aligned RoPE for Mixed-Resolution Diffusion Transformer

    cs.CV 2025-11 conditional novelty 6.0 of 10

    Expressing all RoPE positions on the query's grid ('one attention, one scale') plus a small boundary content-exchange step restores mixed-resolution diffusion generation that naive position interpolation destroys.

  3. Pusa V1.0: Unlocking Temporal Control in Pretrained Video Diffusion Models via Vectorized Timestep Adaptation

    cs.CV 2025-07 conditional novelty 6.0 of 10

    A $500 LoRA fine-tune of Wan2.1-T2V with per-frame random timesteps matches Wan-I2V's benchmark quality and adds zero-shot start-end and video-extension capabilities.

  4. LoViC: Efficient Long Video Generation with Context Compression

    cs.CV 2025-07 conditional novelty 6.0 of 10

    LoViC uses FlexFormer, a single-query-token Q-Former with interpolated rotary positional encoding, to compress long video-text context for efficient long-video generation.

  5. FreeLong++: Training-Free Long Video Generation via Multi-band SpectralFusion

    cs.CV 2025-06 conditional novelty 6.0 of 10

    FreeLong++ extends short-video diffusion models to 4x to 8x longer clips, without retraining, by fusing multiple windowed attention branches through frequency-domain filters and a spectral noise initialization.

  6. FastInit: Fast Noise Initialization for Temporally Consistent Video Generation

    cs.CV 2025-06 conditional novelty 6.0 of 10

    A single-pass learned noise predictor, trained to imitate FreeInit's outputs, gives temporally more consistent text-to-video generation at near-zero added inference cost.

  7. Visko Orbis 1.0: A Live Model for Real-Time Interactive Long Video Generation

    cs.CV 2026-07 conditional novelty 5.0 of 10

    A chunk-wise streaming video model with bounded multi-scale memory and streaming 4K upscaling reports real-time interactive long-form generation and top Arena preference/stability scores.

  8. BIFE: Better Interaction, Fewer Errors for Minute-Long Video Generation

    cs.CV 2025-11 conditional novelty 5.0 of 10

    BlockVid generates minute-long videos with a semantic sparse KV cache, Block Forcing training, and chunk-level noise scheduling, reporting large gains on its own LV-Bench and on VBench.

  9. LongDWM: Cross-Granularity Distillation for Building a Long-Term Driving World Model

    cs.CV 2025-06 reject novelty 5.0 of 10

    A hierarchical coarse-to-fine diffusion transformer with cross-granularity distillation improves long-term driving video prediction, but the reported gains may be inflated by future-derived text prompts and a selected...

  10. LongVie: Multimodal-Guided Controllable Ultra-Long Video Generation

    cs.CV 2025-08 conditional novelty 4.0 of 10

    LongVie combines unified noise initialization, global control normalization, and multi-modal depth-plus-keypoint guidance to generate temporally consistent controllable videos of up to one minute.

Pith tools