Recent advancements in machine learning, particularly in neural networks and deep learning, have significantly enhanced Natural Language Processing (NLP) capabilities, especially in complex tasks like document-level classification used in legal text analysis. However, while models such as BERT and GPT-4 have proven effective, they are often criticized for their lack of interpretability, posing challenges in domains like law where transparency is crucial. This research project aims to address this issue by developing eXplainable Artificial Intelligence (XAI) approaches tailored for long-text classification in the legal domain, enhancing the transparency and reliability of these models for real-world decision-making applications.