Towards Grad-CAM Based Explainability in a Legal Text Processing Pipeline
This work addresses the critical need for model transparency and interpretability in legal AI systems, which is crucial for legal professionals and stakeholders to understand decision-making processes. It is an incremental step towards explainability for legal text processing.
This paper explores the application of Grad-CAM, an image processing technique, to legal texts to enhance explainability in legal decision-making models. It demonstrates how adapted Grad-CAM metrics can reveal the impact of embedding choices and contextual information on downstream processing in legal text analysis.
Explainable AI(XAI)is a domain focused on providing interpretability and explainability of a decision-making process. In the domain of law, in addition to system and data transparency, it also requires the (legal-) decision-model transparency and the ability to understand the models inner working when arriving at the decision. This paper provides the first approaches to using a popular image processing technique, Grad-CAM, to showcase the explainability concept for legal texts. With the help of adapted Grad-CAM metrics, we show the interplay between the choice of embeddings, its consideration of contextual information, and their effect on downstream processing.