# Sparse and Structured Visual Attention

Source: [Sparse and Structured Visual Attention](<https://www.sota2.com/research/paper/68fa6a591e209131cea27a37>)

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

Data updated: 2025-10-25T23:22:37.170Z

Authors: Pedro Henrique Martins, Vlad Niculae, Zita Marinho, Andr\\'e Martins

Published: 2020-02-13T15:08:12.000Z

## Abstract

Visual attention mechanisms are widely used in multimodal tasks, as visual question answering (VQA). One drawback of softmax-based attention mechanisms is that they assign some probability mass to all image regions, regardless of their adjacency structure and of their relevance to the text. In this paper, to better link the image structure with the text, we replace the traditional softmax attention mechanism with two alternative sparsity-promoting transformations: sparsemax, which is able to select only the relevant regions (assigning zero weight to the rest), and a newly proposed Total-Variation Sparse Attention (TVmax), which further encourages the joint selection of adjacent spatial locations. Experiments in VQA show gains in accuracy as well as higher similarity to human attention, which suggests better interpretability.

## 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/2002.05556>)

[PDF](<https://arxiv.org/pdf/2002.05556.pdf>)
