Pith. sign in

REVIEW 25 cited by

Fast Training of Diffusion Models with Masked Transformers

Not yet reviewed by Pith; the record is open.

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 2306.09305 v2 pith:A7BULS5I submitted 2023-06-15 cs.CV cs.AIcs.LG

Fast Training of Diffusion Models with Masked Transformers

classification cs.CV cs.AIcs.LG
keywords trainingmaskedpatchesdiffusionmodelsgenerativetransformertransformers
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

We propose an efficient approach to train large diffusion models with masked transformers. While masked transformers have been extensively explored for representation learning, their application to generative learning is less explored in the vision domain. Our work is the first to exploit masked training to reduce the training cost of diffusion models significantly. Specifically, we randomly mask out a high proportion (e.g., 50%) of patches in diffused input images during training. For masked training, we introduce an asymmetric encoder-decoder architecture consisting of a transformer encoder that operates only on unmasked patches and a lightweight transformer decoder on full patches. To promote a long-range understanding of full patches, we add an auxiliary task of reconstructing masked patches to the denoising score matching objective that learns the score of unmasked patches. Experiments on ImageNet-256x256 and ImageNet-512x512 show that our approach achieves competitive and even better generative performance than the state-of-the-art Diffusion Transformer (DiT) model, using only around 30% of its original training time. Thus, our method shows a promising way of efficiently training large transformer-based diffusion models without sacrificing the generative performance.

discussion (0)

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

Forward citations

Cited by 25 Pith papers

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

  1. MaskAlign: Token-Subset Representation Alignment for Efficient Diffusion Training

    cs.CV 2026-06 unverdicted novelty 7.0

    MaskAlign uses random token-subset alignment and pre-mask mixing to reduce diffusion models' reliance on complete clean-image token sets during representation alignment.

  2. Language Model Beats Diffusion -- Tokenizer is Key to Visual Generation

    cs.CV 2023-10 unverdicted novelty 7.0

    A new shared video-image tokenizer enables large language models to surpass diffusion models on standard visual generation benchmarks.

  3. PixelU: A U-Shaped Transformer for Efficient End-to-End Pixel Diffusion

    cs.CV 2026-06 unverdicted novelty 6.0

    PixelU is a minimalist U-shaped Diffusion Transformer for pixel-space diffusion that decouples frequencies with zero-cost skip connections and constant-channel downsampling, outperforming baselines like JiT-G at 1/3 t...

  4. DiffusionBench: On Holistic Evaluation of Diffusion Transformers

    cs.CV 2026-06 conditional novelty 6.0

    NanoGen unifies DiT training on ImageNet and T2I, reveals negative Pearson correlations (-0.377 to -0.580) in method rankings across metrics from 21 models, and motivates DiffusionBench for holistic evaluation.

  5. IDEAL: In-DEpth ALignment Makes A Discrete Representation AutoEncoder

    cs.CV 2026-06 unverdicted novelty 6.0

    IDEAL improves discrete representation autoencoders by jointly aligning quantized tokens with shallow and deep VFM features, reporting 0.61 rFID on ImageNet and 1.89 gFID for autoregressive image generation.

  6. Vision Foundation Models as Generalist Tokenizers for Image Generation

    cs.CV 2026-05 unverdicted novelty 6.0

    VFMTok builds a generalist image tokenizer on frozen VFMs using adaptive quantization and semantic alignment, delivering gFID 1.36 for autoregressive and 1.25 for continuous generation on ImageNet with 3x faster convergence.

  7. SRC-Flow: Compact Semantic Representations Enable Normalizing Flows for Image Generation

    cs.CV 2026-05 conditional novelty 6.0

    A learned 32-channel semantic compression lets a normalizing flow reach gFID 1.65 on ImageNet 256×256, the best FID reported for flow-based image generation.

  8. SRC-Flow: Compact Semantic Representations Enable Normalizing Flows for Image Generation

    cs.CV 2026-05 unverdicted novelty 6.0

    SRC-Flow compresses RAE features via a Semantic Representation Compressor into a low-dimensional space, enabling normalizing flows to reach gFID 1.65 on ImageNet 256x256 and 2.07 on 512x512 while retaining exact likelihoods.

  9. SRC-Flow: Compact Semantic Representations Enable Normalizing Flows for Image Generation

    cs.CV 2026-05 unverdicted novelty 6.0

    SRC-Flow compresses RAE features into a low-dimensional semantic space with a Semantic Representation Compressor, enabling normalizing flows to achieve SOTA gFID scores of 1.65 and 2.07 on ImageNet 256x256 and 512x512...

  10. Beyond Point-Wise Matching: Structural Representation Alignment for Accelerating Diffusion Transformers

    cs.CV 2026-05 unverdicted novelty 6.0

    sREPA enforces structural consistency in relational geometry of pre-trained vision features to accelerate DiT training and improve generation quality.

  11. End-to-End Autoregressive Image Generation with 1D Semantic Tokenizer

    cs.CV 2026-05 unverdicted novelty 6.0

    An end-to-end autoregressive model with a jointly trained 1D semantic tokenizer achieves state-of-the-art FID 1.48 on ImageNet 256x256 generation without guidance.

  12. Denoising, Fast and Slow: Difficulty-Aware Adaptive Sampling for Image Generation

    cs.CV 2026-04 unverdicted novelty 6.0

    Patch Forcing enables diffusion models to denoise image patches at varying rates based on predicted difficulty, advancing easier regions first to improve context and achieve better generation quality on ImageNet while...

  13. ELT: Elastic Looped Transformers for Visual Generation

    cs.CV 2026-04 conditional novelty 6.0

    Weight-shared looped transformers trained with intra-loop self-distillation match MaskGIT-class FID/FVD at roughly 4x fewer parameters and support any-time inference across loop counts.

  14. ELT: Elastic Looped Transformers for Visual Generation

    cs.CV 2026-04 unverdicted novelty 6.0

    Elastic Looped Transformers share weights across recurrent blocks and apply intra-loop self-distillation to deliver 4x parameter reduction while matching competitive FID and FVD scores on ImageNet and UCF-101.

  15. Data Warmup: Complexity-Aware Curricula for Efficient Diffusion Training

    cs.LG 2026-04 conditional novelty 6.0

    Data Warmup accelerates diffusion training on ImageNet by scheduling images from low to high complexity via a foreground-based metric and temperature-controlled sampler, improving FID and IS scores faster than uniform...

  16. Prompt Reinjection: Alleviating Prompt Forgetting in Multimodal Diffusion Transformers for Text-to-Image Generation

    cs.CV 2026-02 conditional novelty 6.0

    Re-injecting shallow text features into deeper MMDiT blocks counteracts measured 'prompt forgetting' and improves instruction following in SD3, SD3.5, FLUX, and Qwen-Image without retraining.

  17. MixFlow Training: Alleviating Exposure Bias with Slowed Interpolation Mixture

    cs.CV 2025-12 conditional novelty 6.0

    Training diffusion models on a mixture of higher-noise interpolations (MixFlow) improves generation FID across SiT, REPA, RAE and SD3.5, reaching ImageNet 256 gFID 1.43 after post-training.

  18. REGLUE Your Latents with Global and Local Semantics for Entangled Diffusion

    cs.CV 2025-12 conditional novelty 6.0

    Nonlinear multi-layer compression of frozen VFM patch semantics, jointly denoised with VAE latents, improves ImageNet 256x256 FID (12.9 vs 15.2 for REG at SiT-B/2, 400K) and accelerates convergence over REPA/ReDi/REG.

  19. VFM-VAE: Vision Foundation Models Can Be Good Tokenizers for Latent Diffusion Models

    cs.CV 2025-10 unverdicted novelty 6.0

    VFM-VAE uses a frozen VFM directly as LDM tokenizer via a custom decoder, reaching gFID 2.22 in 80 epochs and 1.62 after 640 epochs.

  20. Diagnosing and Improving Diffusion Models by Estimating the Optimal Loss Value

    cs.LG 2025-06 conditional novelty 6.0

    Derives closed-form optimal loss for unified diffusion models, provides variance-controlled estimators, and shows improved diagnosis, training schedules, and power-law scaling after subtracting the optimal value.

  21. From SRA to Self-Flow: Data Augmentation or Self-Supervision?

    cs.CV 2026-07 unverdicted novelty 5.0

    Attention Separation ablations show that gains from SRA to Self-Flow in diffusion transformers arise mainly from noise-dimension data augmentation rather than token-level self-supervision.

  22. FrequencyBooster: Full-Frequency Modeling for High-Fidelity Pixel Diffusion

    cs.CV 2026-05 unverdicted novelty 5.0

    FrequencyBooster reports state-of-the-art FID scores of 1.60 at 256x256 and 1.69 at 512x512 for pixel diffusion by using a specialized decoder for full-frequency modeling.

  23. Learning on the Manifold: Unlocking Standard Diffusion Transformers with Representation Encoders

    cs.LG 2026-02 conditional novelty 5.0

    Training flow matching along sphere geodesics with a curvature-aware loss weight lets standard DiT-B converge on DINOv2 features (FID 3.37 with guidance), contradicting the need for width scaling.

  24. Prompt Reinjection: Alleviating Prompt Forgetting in Multimodal Diffusion Transformers for Text-to-Image Generation

    cs.CV 2026-02 unverdicted novelty 5.0

    Prompt reinjection alleviates progressive forgetting of text prompt semantics in MMDiT text branches, producing consistent improvements in text-to-image instruction following on GenEval, DPG, and T2I-CompBench++.

  25. Elucidating Representation Degradation Problem in Diffusion Model Training

    cs.LG 2026-05 unverdicted novelty 4.0

    Diffusion models suffer representation degradation at high noise due to recoverability mismatch; ERD mitigates this by dynamic optimization reallocation, accelerating convergence across backbones.