A-SelecT: Automatic Timestep Selection for Diffusion Transformer Representation Learning

Read full text Summarise with VIRAT
Authors: Changyu Liu, Jia Liang, Wenhao Yang, Yiming Cui, Jinghao Yang, Tianyang Wang, Qifan Wang, Dongfang Liu, Cheng Han
Cited by: 0
References: 71

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

Diffusion models have significantly reshaped the field of generative artificial intelligence and are now increasingly explored for their capacity in discriminative representation learning. Diffusion Transformer (DiT) has recently gained attention as a promising alternative to conventional U-Net-based diffusion models, demonstrating a promising avenue for downstream discriminative tasks via generative pre-training. However, its current training efficiency and representational capacity remain largely constrained due to the inadequate timestep searching and insufficient exploitation of DiT-specific feature representations. In light of this view, we introduce Automatically Selected Timestep (A-SelecT) that dynamically pinpoints DiT's most information-rich timestep from the selected transformer feature in a single run, eliminating the need for both computationally intensive exhaustive timestep searching and suboptimal discriminative feature selection. Extensive experiments on classification and segmentation benchmarks demonstrate that DiT, empowered by A-SelecT, surpasses all prior diffusion-based attempts efficiently and effectively.

Fields of Study & Subjects

Generative Adversarial Networks and Image Synthesis Advanced Neural Network Applications Domain Adaptation and Few-Shot Learning

Publication details

Journal / source
arXiv (Cornell University)
Publication date
25 March 2026
Language
EN
Document type
Preprint