Why Are You Wrong? Counterfactual Explanations for Language Grounding with 3D Objects

TL;DR

This paper introduces a method to generate counterfactual explanations for neural network models in 3D object grounding, helping to understand and improve model behavior in language-based robotic and design applications.

Contribution

It presents a novel approach to generate meaningful counterfactual examples that reveal model weaknesses and biases in 3D object grounding tasks using natural language descriptions.

Findings

Counterfactual examples expose model biases and weaknesses.

The method improves understanding of model behavior.

Evaluation on ShapeTalk dataset demonstrates effectiveness.

Abstract

Combining natural language and geometric shapes is an emerging research area with multiple applications in robotics and language-assisted design. A crucial task in this domain is object referent identification, which involves selecting a 3D object given a textual description of the target. Variability in language descriptions and spatial relationships of 3D objects makes this a complex task, increasing the need to better understand the behavior of neural network models in this domain. However, limited research has been conducted in this area. Specifically, when a model makes an incorrect prediction despite being provided with a seemingly correct object description, practitioners are left wondering: "Why is the model wrong?". In this work, we present a method answering this question by generating counterfactual examples. Our method takes a misclassified sample, which includes two objects and a text description, and generates an alternative yet similar formulation that would have resulted in a correct prediction by the model. We have evaluated our approach with data from the ShapeTalk dataset along with three distinct models. Our counterfactual examples maintain the structure of the original description, are semantically similar and meaningful. They reveal weaknesses in the description, model bias and enhance the understanding of the models behavior. Theses insights help practitioners to better interact with systems as well as engineers to improve models.