CVJun 8, 2023

Multi-Scale And Token Mergence: Make Your ViT More Efficient

arXiv:2306.04897v28 citationsh-index: 10
Originality Incremental advance
AI Analysis

This addresses efficiency issues in Vision Transformers for computer vision applications, representing an incremental improvement over existing token pruning techniques.

The paper tackles the computational expense of Vision Transformers by proposing a token pruning method that merges non-crucial tokens with crucial ones to retain information, achieving a 33% reduction in computational costs with only a 0.1% accuracy drop on DeiT-S for ImageNet.

Since its inception, Vision Transformer (ViT) has emerged as a prevalent model in the computer vision domain. Nonetheless, the multi-head self-attention (MHSA) mechanism in ViT is computationally expensive due to its calculation of relationships among all tokens. Although some techniques mitigate computational overhead by discarding tokens, this also results in the loss of potential information from those tokens. To tackle these issues, we propose a novel token pruning method that retains information from non-crucial tokens by merging them with more crucial tokens, thereby mitigating the impact of pruning on model performance. Crucial and non-crucial tokens are identified by their importance scores and merged based on similarity scores. Furthermore, multi-scale features are exploited to represent images, which are fused prior to token pruning to produce richer feature representations. Importantly, our method can be seamlessly integrated with various ViTs, enhancing their adaptability. Experimental evidence substantiates the efficacy of our approach in reducing the influence of token pruning on model performance. For instance, on the ImageNet dataset, it achieves a remarkable 33% reduction in computational costs while only incurring a 0.1% decrease in accuracy on DeiT-S.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes