EC-DIT: Scaling Diffusion Transformers with Adaptive Expert-Choice Routing
AuthorsHaotian Sun†‡, Tao Lei, Bowen Zhang, Yanghao Li, Haoshuo Huang, Ruoming Pang, Bo Dai‡, Nan Du
AuthorsHaotian Sun†‡, Tao Lei, Bowen Zhang, Yanghao Li, Haoshuo Huang, Ruoming Pang, Bo Dai‡, Nan Du
Diffusion transformers have been widely adopted for text-to-image synthesis. While scaling these models up to billions of parameters shows promise, the effectiveness of scaling beyond current sizes remains underexplored and challenging. By explicitly exploiting the computational heterogeneity of image generations, we develop a new family of Mixture-of-Experts (MoE) models (EC-DIT) for diffusion transformers with expert-choice routing. EC-DIT learns to adaptively optimize the compute allocated to understand the input texts and generate the respective image patches, enabling heterogeneous computation aligned with varying text-image complexities. This heterogeneity provides an efficient way of scaling EC-DIT up to 97 billion parameters and achieving significant improvements in training convergence, text-to-image alignment, and overall generation quality over dense models and conventional MoE models. Through extensive ablations, we show that EC-DIT demonstrates superior scalability and adaptive compute allocation by recognizing varying textual importance through end-to-end training. Notably, in text-to-image alignment evaluation, our largest models achieve a state-of-the-art GenEval score of 71.68% and still maintain competitive inference speed with intuitive interpretability.
†Work done during an Apple internship.
‡Georgia Institute of Technology
July 11, 2025research area Methods and Algorithms, research area Speech and Natural Language Processing
Mixture-of-experts (MoE) architectures have expanded from language modeling to automatic speech recognition (ASR). Traditional MoE methods, such as the Switch Transformer, route experts independently within each layer. Our analysis reveals that routers in most layers make expert choices that are not strongly correlated with the choices of the routers in other layers. To increase the cooperation between experts in different layers and encourage...
November 18, 2024research area Speech and Natural Language ProcessingWorkshop at NeurIPS
This paper was accepted at the Efficient Natural Language and Speech Processing (ENLSP) Workshop at NeurIPS 2024.
Large Language Models (LLMs) typically generate outputs token by token using a fixed compute budget, leading to inefficient resource utilization. To address this shortcoming, recent advancements in mixture of expert (MoE) models, speculative decoding, and early exit strategies leverage the insight that computational demands can vary...