← Previous · All Episodes · Next →
Mixture-of-Transformers: A Sparse and Scalable Architecture for Multi-Modal Foundation Models Episode 51

Mixture-of-Transformers: A Sparse and Scalable Architecture for Multi-Modal Foundation Models

· 24:52

|

🤗 Paper Upvotes: 25 | cs.CL

Authors:
Weixin Liang, Lili Yu, Liang Luo, Srinivasan Iyer, Ning Dong, Chunting Zhou, Gargi Ghosh, Mike Lewis, Wen-tau Yih, Luke Zettlemoyer, Xi Victoria Lin

Title:
Mixture-of-Transformers: A Sparse and Scalable Architecture for Multi-Modal Foundation Models

Arxiv:
http://arxiv.org/abs/2411.04996v1

Abstract:
The development of large language models (LLMs) has expanded to multi-modal systems capable of processing text, images, and speech within a unified framework. Training these models demands significantly larger datasets and computational resources compared to text-only LLMs. To address the scaling challenges, we introduce Mixture-of-Transformers (MoT), a sparse multi-modal transformer architecture that significantly reduces pretraining computational costs. MoT decouples non-embedding parameters of the model by modality -- including feed-forward networks, attention matrices, and layer normalization -- enabling modality-specific processing with global self-attention over the full input sequence. We evaluate MoT across multiple settings and model scales. In the Chameleon 7B setting (autoregressive text-and-image generation), MoT matches the dense baseline's performance using only 55.8\% of the FLOPs. When extended to include speech, MoT reaches speech performance comparable to the dense baseline with only 37.2\% of the FLOPs. In the Transfusion setting, where text and image are trained with different objectives, a 7B MoT model matches the image modality performance of the dense baseline with one third of the FLOPs, and a 760M MoT model outperforms a 1.4B dense baseline across key image generation metrics. System profiling further highlights MoT's practical benefits, achieving dense baseline image quality in 47.2\% of the wall-clock time and text quality in 75.6\% of the wall-clock time (measured on AWS p4de.24xlarge instances with NVIDIA A100 GPUs).


Subscribe

Listen to Daily Paper Cast using one of many popular podcasting apps or directories.

Apple Podcasts Spotify Overcast Pocket Casts
← Previous · All Episodes · Next →