# DeBiFormer: Vision Transformer with Deformable Agent Bi-level Routing Attention

Source: [DeBiFormer: Vision Transformer with Deformable Agent Bi-level Routing Attention](<https://www.sota2.com/research/paper/68fa75351e209131cea666e6>)

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

Data updated: 2026-01-04T04:03:23.374Z

Authors: Nguyen Huu Bao Long, Chenyu Zhang, Yuzhi Shi, Tsubasa Hirakawa, Takayoshi Yamashita, Tohgoroh Matsui, Hironobu Fujiyoshi

Published: 2024-10-11T07:23:10.000Z

## Abstract

Vision Transformers with various attention modules have demonstrated superior performance on vision tasks. While using sparsity-adaptive attention, such as in DAT, has yielded strong results in image classification, the key-value pairs selected by deformable points lack semantic relevance when fine-tuning for semantic segmentation tasks. The query-aware sparsity attention in BiFormer seeks to focus each query on top-k routed regions. However, during attention calculation, the selected key-value pairs are influenced by too many irrelevant queries, reducing attention on the more important ones. To address these issues, we propose the Deformable Bi-level Routing Attention (DBRA) module, which optimizes the selection of key-value pairs using agent queries and enhances the interpretability of queries in attention maps. Based on this, we introduce the Deformable Bi-level Routing Attention Transformer (DeBiFormer), a novel general-purpose vision transformer built with the DBRA module. DeBiFormer has been validated on various computer vision tasks, including image classification, object detection, and semantic segmentation, providing strong evidence of its effectiveness.Code is available at {https://github.com/maclong01/DeBiFormer}

## 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.

- [Instance Segmentation on COCO 2017](<https://www.sota2.com/research/sota/instance-segmentation-on-coco-2017.md>): Instance Segmentation; COCO 2017; APm: 43.2.

- [Object Detection on COCO 2017](<https://www.sota2.com/research/sota/object-detection-on-coco-2017.md>): Object Detection; COCO 2017; AP (Box): 47.1.

- [Semantic Segmentation on ADE20K (val)](<https://www.sota2.com/research/sota/semantic-segmentation-on-ade20k-val.md>): Semantic segmentation; ADE20K (val); mIoU: 51.4.

- [Image Classification on ImageNet-1K 1.0 (val)](<https://www.sota2.com/research/sota/image-classification-on-imagenet-1k-1-0-val.md>): Image Classification; ImageNet-1K 1.0 (val); Top-1 Accuracy: 84.4.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2410.08582>)

[PDF](<https://arxiv.org/pdf/2410.08582.pdf>)

[Implementation](<https://github.com/maclong01/DeBiFormer>)