# High-Fidelity Differential-information Driven Binary Vision Transformer

Source: [High-Fidelity Differential-information Driven Binary Vision Transformer](<https://www.sota2.com/research/paper/68fa79fe1e209131cea80894>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-27T06:57:50.280Z

Authors: Tian Gao, Zhiyuan Zhang, Kaijie Yin, Xu-Cheng Zhong, Hui Kong

Published: 2025-07-03T00:59:53.000Z

## Abstract

The binarization of vision transformers (ViTs) offers a promising approach to addressing the trade-off between high computational/storage demands and the constraints of edge-device deployment. However, existing binary ViT methods often suffer from severe performance degradation or rely heavily on full-precision modules. To address these issues, we propose DIDB-ViT, a novel binary ViT that is highly informative while maintaining the original ViT architecture and computational efficiency. Specifically, we design an informative attention module incorporating differential information to mitigate information loss caused by binarization and enhance high-frequency retention. To preserve the fidelity of the similarity calculations between binary Q and K tensors, we apply frequency decomposition using the discrete Haar wavelet and integrate similarities across different frequencies. Additionally, we introduce an improved RPReLU activation function to restructure the activation distribution, expanding the model's representational capacity. Experimental results demonstrate that our DIDB-ViT significantly outperforms state-of-the-art network quantization methods in multiple ViT architectures, achieving superior image classification and segmentation performance.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2507.02222>)

[PDF](<https://arxiv.org/pdf/2507.02222.pdf>)
