Reconciling feature sharing and multiple predictions with MIMO Vision Transformers | Remy Sun

Published by

remysun.github.io

on •1 min read

Reconciling feature sharing and multiple predictions with MIMO Vision Transformers | Remy Sun

Multi-input multi-output training improves network performance by optimizing multiple subnetworks simultaneously. In this paper, we propose MixViT, the first MIMO framework for vision transformers that takes advantage of ViTs’ innate mechanisms to share features between subnetworks.

Reconciling Feature Sharing Multiple Predictions

Learn more