New depthwise convolution speeds up vision foundation models

Researchers have developed a new method to speed up vision foundation models by replacing certain attention heads in Vision Transformer (ViT) backbones with efficient depthwise convolution layers. This drop-in replacement achieves a 17-20% inference speedup with minimal performance loss on image classification and segmentation tasks. The approach includes strategies for identifying replaceable heads and a fine-tuning procedure to restore downstream task performance, with a reference implementation made publicly available. AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

IMPACT Accelerates inference for vision foundation models, potentially enabling wider deployment on resource-constrained devices.

RANK_REASON The cluster contains an academic paper detailing a new technical approach for accelerating existing models. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• arXiv
• depthwise convolution
• Vision Transformer

• paper
• infra