Attentive Explanations: Justifying Decisions and Pointing to the Evidence

TL;DR

This paper introduces the PJ-X model that generates natural language explanations and points to evidence in images for visual decision tasks, aiming to make deep models more interpretable and human-like.

Contribution

The paper presents a novel model capable of providing both textual justifications and visual evidence, along with new datasets for explainable visual decision making.

Findings

PJ-X outperforms prior models in explanation quality

The model effectively points to relevant evidence in images

Human evaluations favor PJ-X explanations

Abstract

Deep models are the defacto standard in visual decision models due to their impressive performance on a wide array of visual tasks. However, they are frequently seen as opaque and are unable to explain their decisions. In contrast, humans can justify their decisions with natural language and point to the evidence in the visual world which led to their decisions. We postulate that deep models can do this as well and propose our Pointing and Justification (PJ-X) model which can justify its decision with a sentence and point to the evidence by introspecting its decision and explanation process using an attention mechanism. Unfortunately there is no dataset available with reference explanations for visual decision making. We thus collect two datasets in two domains where it is interesting and challenging to explain decisions. First, we extend the visual question answering task to not only provide an answer but also a natural language explanation for the answer. Second, we focus on explaining human activities which is traditionally more challenging than object classification. We extensively evaluate our PJ-X model, both on the justification and pointing tasks, by comparing it to prior models and ablations using both automatic and human evaluations.