# SD-DiT: Unleashing the Power of Self-supervised Discrimination in Diffusion Transformer

arXiv: https://arxiv.org/abs/2403.17004 · HTML: https://mlsanity.com/paper/2403.17004/

## Key facts

- Cited by 49 papers in the map (data through 14 Oct 2026).
- Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models.
- Builds on 76 of its 83 references that are in the map.
- First posted on arXiv 25 Mar 2024.

## Placed in

- Diffusion models / [Diffusion foundations and sampling](https://mlsanity.com/diffusion/diffusion-foundations-and-sampling/)
- Knowledge distillation / [Multimodal distillation](https://mlsanity.com/distill/multimodal-distillation/)

## Builds on (most cited first, 30 of 76)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [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)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Deep Unsupervised Learning using Nonequilibrium Thermodynamics](https://mlsanity.com/paper/1503.03585/) (2015)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding](https://mlsanity.com/paper/2205.11487/) (2022)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [Bootstrap your own latent: A new approach to self-supervised Learning](https://mlsanity.com/paper/2006.07733/) (2020)
- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [Generative Modeling by Estimating Gradients of the Data Distribution](https://mlsanity.com/paper/1907.05600/) (2019)
- [Improved Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2102.09672/) (2021)
- [GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models](https://mlsanity.com/paper/2112.10741/) (2021)
- [DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation](https://mlsanity.com/paper/2208.12242/) (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)
- [Unsupervised Learning of Visual Features by Contrasting Cluster Assignments](https://mlsanity.com/paper/2006.09882/) (2020)
- [Large Scale GAN Training for High Fidelity Natural Image Synthesis](https://mlsanity.com/paper/1809.11096/) (2018)

## Built on by (most cited first, 30 of 49)

- [Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think](https://mlsanity.com/paper/2410.06940/) (2024)
- [Diffusion Transformers with Representation Autoencoders](https://mlsanity.com/paper/2510.11690/) (2025)
- [Representation Entanglement for Generation: Training Diffusion Transformers Is Much Easier Than You Think](https://mlsanity.com/paper/2507.01467/) (2025)
- [EQ-VAE: Equivariance Regularized Latent Space for Improved Generative Image Modeling](https://mlsanity.com/paper/2502.09509/) (2025)
- [FasterDiT: Towards Faster Diffusion Transformers Training without Architecture Modification](https://mlsanity.com/paper/2410.10356/) (2024)
- [No Other Representation Component Is Needed: Diffusion Transformers Can Provide Representation Guidance by Themselves](https://mlsanity.com/paper/2505.02831/) (2025)
- [Boosting Generative Image Modeling via Joint Image-Feature Synthesis](https://mlsanity.com/paper/2504.16064/) (2025)
- [HiDream-O1-Image: A Natively Unified Image Generative Foundation Model with Pixel-level Unified Transformer](https://mlsanity.com/paper/2605.11061/) (2026)
- [Self-Supervised Flow Matching for Scalable Multi-Modal Synthesis](https://mlsanity.com/paper/2603.06507/) (2026)
- [MaskGWM: A Generalizable Driving World Model with Video Mask Reconstruction](https://mlsanity.com/paper/2502.11663/) (2025)
- [Hi3D: Pursuing High-Resolution Image-to-3D Generation with Video Diffusion Models](https://mlsanity.com/paper/2409.07452/) (2024)
- [Region-Constraint In-Context Generation for Instructional Video Editing](https://mlsanity.com/paper/2512.17650/) (2025)
- [ResMaster: Mastering High-Resolution Image Generation via Structural and Fine-Grained Guidance](https://mlsanity.com/paper/2406.16476/) (2024)
- [D-OPSD: On-Policy Self-Distillation for Continuously Tuning Step-Distilled Diffusion Models](https://mlsanity.com/paper/2605.05204/) (2026)
- [TREAD: Token Routing for Efficient Architecture-agnostic Diffusion Training](https://mlsanity.com/paper/2501.04765/) (2025)
- [Guiding a Diffusion Transformer with the Internal Dynamics of Itself](https://mlsanity.com/paper/2512.24176/) (2025)
- [Improving Virtual Try-On with Garment-focused Diffusion Models](https://mlsanity.com/paper/2409.08258/) (2024)
- [Learning Diffusion Models with Flexible Representation Guidance](https://mlsanity.com/paper/2507.08980/) (2025)
- [Improving Text-guided Object Inpainting with Semantic Pre-inpainting](https://mlsanity.com/paper/2409.08260/) (2024)
- [SARA: Structural and Adversarial Representation Alignment for Training-efficient Diffusion Models](https://mlsanity.com/paper/2503.08253/) (2025)
- [LayerSync: Self-aligning Intermediate Layers](https://mlsanity.com/paper/2510.12581/) (2025)
- [SRA 2: Variational Autoencoder Self-Representation Alignment for Efficient Diffusion Training](https://mlsanity.com/paper/2601.17830/) (2026)
- [Incorporating Visual Correspondence into Diffusion Model for Virtual Try-On](https://mlsanity.com/paper/2505.16977/) (2025)
- [No Alignment Needed for Generation: Learning Linearly Separable Representations in Diffusion Models](https://mlsanity.com/paper/2509.21565/) (2025)
- [FreeEnhance: Tuning-Free Image Enhancement via Content-Consistent Noising-and-Denoising Process](https://mlsanity.com/paper/2409.07451/) (2024)
- [Efficient Generative Model Training via Embedded Representation Warmup](https://mlsanity.com/paper/2504.10188/) (2025)
- [EDT: An Efficient Diffusion Transformer Framework Inspired by Human-like Sketching](https://mlsanity.com/paper/2410.23788/) (2024)
- [REGLUE Your Latents with Global and Local Semantics for Entangled Diffusion](https://mlsanity.com/paper/2512.16636/) (2025)
- [DiverseDiT: Towards Diverse Representation Learning in Diffusion Transformers](https://mlsanity.com/paper/2603.04239/) (2026)
- [Denoising, Fast and Slow: Difficulty-Aware Adaptive Sampling for Image Generation](https://mlsanity.com/paper/2604.19141/) (2026)

## Abstract (arXiv)

Diffusion Transformer (DiT) has emerged as the new trend of generative diffusion models on image generation. In view of extremely slow convergence in typical DiT, recent breakthroughs have been driven by mask strategy that significantly improves the training efficiency of DiT with additional intra-image contextual learning. Despite this progress, mask strategy still suffers from two inherent limitations: (a) training-inference discrepancy and (b) fuzzy relations between mask reconstruction & generative diffusion process, resulting in sub-optimal training of DiT. In this work, we address these limitations by novelly unleashing the self-supervised discrimination knowledge to boost DiT training. Technically, we frame our DiT in a teacher-student manner. The teacher-student discriminative pairs are built on the diffusion noises along the same Probability Flow Ordinary Differential Equation (PF-ODE). Instead of applying mask reconstruction loss over both DiT encoder and decoder, we decouple DiT encoder and decoder to separately tackle discriminative and generative objectives. In particular, by encoding discriminative pairs with student and teacher DiT encoders, a new discriminative loss is designed to encourage the inter-image alignment in the self-supervised embedding space. After that, student samples are fed into student DiT decoder to perform the typical generative diffusion task. Extensive experiments are conducted on ImageNet dataset, and our method achieves a competitive balance between training cost and generative capacity.