Where is the Model Looking At?--Concentrate and Explain the Network Attention

TL;DR

This paper introduces an explainable multi-task framework that enhances model interpretability and attention focus on discriminative image regions, improving trust and performance across various models and datasets.

Contribution

The paper proposes the EAT framework that integrates attribute prediction with multi-task learning to produce interpretable, multi-modal explanations and improve model attention and accuracy.

Findings

EAT framework provides effective visual and textual explanations.

Guided attention improves recognition performance.

Framework generalizes across models and datasets.

Abstract

Image classification models have achieved satisfactory performance on many datasets, sometimes even better than human. However, The model attention is unclear since the lack of interpretability. This paper investigates the fidelity and interpretability of model attention. We propose an Explainable Attribute-based Multi-task (EAT) framework to concentrate the model attention on the discriminative image area and make the attention interpretable. We introduce attributes prediction to the multi-task learning network, helping the network to concentrate attention on the foreground objects. We generate attribute-based textual explanations for the network and ground the attributes on the image to show visual explanations. The multi-model explanation can not only improve user trust but also help to find the weakness of network and dataset. Our framework can be generalized to any basic model. We perform experiments on three datasets and five basic models. Results indicate that the EAT framework can give multi-modal explanations that interpret the network decision. The performance of several recognition approaches is improved by guiding network attention.