Pith. sign in

REVIEW 5 cited by

ZigMa: A DiT-style Zigzag Mamba Diffusion Model

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 2403.13802 v3 pith:56OZFSAQ submitted 2024-03-20 cs.CV cs.AIcs.CLcs.LG

Vincent Tao Hu , Stefan Andreas Baumann , Ming Gui , Olga Grebenkova , Pingchuan Ma , Johannes Schusterbauer , Björn Ommer This is my paper · ORCID
classification cs.CVcs.AIcs.CLcs.LG
keywords mambamodelzigzagbaselinesdiffusionlongmamba-basedscalability
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
abstract

The diffusion model has long been plagued by scalability and quadratic complexity issues, especially within transformer-based structures. In this study, we aim to leverage the long sequence modeling capability of a State-Space Model called Mamba to extend its applicability to visual data generation. Firstly, we identify a critical oversight in most current Mamba-based vision methods, namely the lack of consideration for spatial continuity in the scan scheme of Mamba. Secondly, building upon this insight, we introduce a simple, plug-and-play, zero-parameter method named Zigzag Mamba, which outperforms Mamba-based baselines and demonstrates improved speed and memory utilization compared to transformer-based baselines. Lastly, we integrate Zigzag Mamba with the Stochastic Interpolant framework to investigate the scalability of the model on large-resolution visual datasets, such as FacesHQ $1024\times 1024$ and UCF101, MultiModal-CelebA-HQ, and MS COCO $256\times 256$ . Code will be released at https://taohu.me/zigma/

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 5 Pith papers

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

  1. Exploring Diffusion Transformer Designs via Grafting

    cs.LG 2025-06 conditional novelty 6.0 of 10

    Grafting uses activation distillation and lightweight fine-tuning to edit pretrained diffusion transformers into hybrid architectures with near-baseline quality at under 2% pretraining compute.

  2. StyleRWKV: High-Quality and High-Efficiency Style Transfer with RWKV-like Architecture

    cs.CV 2024-12 conditional novelty 6.0 of 10

    StyleRWKV applies recurrent RWKV-style attention with deformable shifting and skip scanning to achieve fast, high-quality arbitrary style transfer.

  3. QuarterMap: Efficient Post-Training Token Pruning for Visual State Space Models

    cs.CV 2025-07 conditional novelty 5.0 of 10

    QuarterMap prunes spatial activations before VMamba's four-directional scan and upsamples after, yielding up to 1.11x throughput with under 1% accuracy loss on ImageNet classification.

  4. MaIR: A Locality- and Continuity-Preserving Mamba for Image Restoration

    cs.CV 2024-12 conditional novelty 5.0 of 10

    MaIR combines a stripe-based S-shaped scanning strategy and a sequence-shuffle attention block to improve Mamba-based image restoration, reporting new best PSNR on 14 benchmarks.

  5. LinGen: Towards High-Resolution Minute-Length Text-to-Video Generation with Linear Computational Complexity

    cs.CV 2024-12 conditional novelty 5.0 of 10

    LinGen replaces self-attention in diffusion transformers with a linear-complexity MATE block, enabling 512p 68-second video generation on a single H100 with quality comparable to Gen-3, LumaLabs, and Kling.

Pith tools