# Autoregressive Distillation of Diffusion Transformers

arXiv: https://arxiv.org/abs/2504.11295 · HTML: https://mlsanity.com/paper/2504.11295/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models.
- Builds on 72 of its 84 references that are in the map.
- First posted on arXiv 15 Apr 2025.

## Placed in

- Diffusion models / [Diffusion foundations and sampling](https://mlsanity.com/diffusion/diffusion-foundations-and-sampling/)

## Builds on (most cited first, 30 of 72)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Score-Based Generative Modeling through Stochastic Differential Equations](https://mlsanity.com/paper/2011.13456/) (2020)
- [Diffusion Models Beat GANs on Image Synthesis](https://mlsanity.com/paper/2105.05233/) (2021)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [Going Deeper with Convolutions](https://mlsanity.com/paper/1409.4842/) (2014)
- [Generative Modeling by Estimating Gradients of the Data Distribution](https://mlsanity.com/paper/1907.05600/) (2019)
- [Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow](https://mlsanity.com/paper/2209.03003/) (2022)
- [Taming Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2012.09841/) (2020)
- [Elucidating the Design Space of Diffusion-Based Generative Models](https://mlsanity.com/paper/2206.00364/) (2022)
- [Improved Techniques for Training GANs](https://mlsanity.com/paper/1606.03498/) (2016)
- [Large Scale GAN Training for High Fidelity Natural Image Synthesis](https://mlsanity.com/paper/1809.11096/) (2018)
- [FiLM: Visual Reasoning with a General Conditioning Layer](https://mlsanity.com/paper/1709.07871/) (2017)
- [DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Steps](https://mlsanity.com/paper/2206.00927/) (2022)
- [SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equations](https://mlsanity.com/paper/2108.01073/) (2021)
- [Consistency Models](https://mlsanity.com/paper/2303.01469/) (2023)
- [Generating Diverse High-Fidelity Images with VQ-VAE-2](https://mlsanity.com/paper/1906.00446/) (2019)
- [Scaling Autoregressive Models for Content-Rich Text-to-Image Generation](https://mlsanity.com/paper/2206.10789/) (2022)
- [MaskGIT: Masked Generative Image Transformer](https://mlsanity.com/paper/2202.04200/) (2022)
- [Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction](https://mlsanity.com/paper/2404.02905/) (2024)
- [One-step Diffusion with Distribution Matching Distillation](https://mlsanity.com/paper/2311.18828/) (2023)
- [Improved Precision and Recall Metric for Assessing Generative Models](https://mlsanity.com/paper/1904.06991/) (2019)
- [PixArt-α: Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesis](https://mlsanity.com/paper/2310.00426/) (2023)
- [Conditional Image Generation with PixelCNN Decoders](https://mlsanity.com/paper/1606.05328/) (2016)

## Built on by (most cited first, 9 of 9)

- [MIDAS: Multimodal Interactive Digital-humAn Synthesis via Real-time Autoregressive Video Generation](https://mlsanity.com/paper/2508.19320/) (2025)
- [SeaCache: Spectral-Evolution-Aware Cache for Accelerating Diffusion Models](https://mlsanity.com/paper/2602.18993/) (2026)
- [Sparse Forcing: Native Trainable Sparse Attention for Real-time Autoregressive Diffusion Video Generation](https://mlsanity.com/paper/2604.21221/) (2026)
- [Distillation of Large Language Models via Concrete Score Matching](https://mlsanity.com/paper/2509.25837/) (2025)
- [ToProVAR: Efficient Visual Autoregressive Modeling via Tri-Dimensional Entropy-Aware Semantic Analysis and Sparsity Optimization](https://mlsanity.com/paper/2602.22948/) (2026)
- [Just-in-Time: Training-Free Spatial Acceleration for Diffusion Transformers](https://mlsanity.com/paper/2603.10744/) (2026)
- [Fast3Dcache: Training-free 3D Geometry Synthesis Acceleration](https://mlsanity.com/paper/2511.22533/) (2025)
- [TASE: Truncation-Aware Semantic Embeddings for 3D Scene Understanding and Editing](https://mlsanity.com/paper/2606.03314/) (2026)
- [XYZFlow:Scaling Multi dimensional Shortcut Flows for Efficient Generative Modeling](https://mlsanity.com/paper/2608.12276/) (2026)

## Abstract (arXiv)

Diffusion models with transformer architectures have demonstrated promising capabilities in generating high-fidelity images and scalability for high resolution. However, iterative sampling process required for synthesis is very resource-intensive. A line of work has focused on distilling solutions to probability flow ODEs into few-step student models. Nevertheless, existing methods have been limited by their reliance on the most recent denoised samples as input, rendering them susceptible to exposure bias. To address this limitation, we propose AutoRegressive Distillation (ARD), a novel approach that leverages the historical trajectory of the ODE to predict future steps. ARD offers two key benefits: 1) it mitigates exposure bias by utilizing a predicted historical trajectory that is less susceptible to accumulated errors, and 2) it leverages the previous history of the ODE trajectory as a more effective source of coarse-grained information. ARD modifies the teacher transformer architecture by adding token-wise time embedding to mark each input from the trajectory history and employs a block-wise causal attention mask for training. Furthermore, incorporating historical inputs only in lower transformer layers enhances performance and efficiency. We validate the effectiveness of ARD in a class-conditioned generation on ImageNet and T2I synthesis. Our model achieves a 5× reduction in FID degradation compared to the baseline methods while requiring only 1.1% extra FLOPs on ImageNet-256. Moreover, ARD reaches FID of 1.84 on ImageNet-256 in merely 4 steps and outperforms the publicly available 1024p text-to-image distilled models in prompt adherence score with a minimal drop in FID compared to the teacher. Project page: https://github.com/alsdudrla10/ARD.