The automatic prediction of court case judgments using Deep Learning and Natural Language Processing is challenged by the variety of norms and regulations, the inherent complexity of the forensic language, and the length of legal judgments. Although state-of-the-art transformer-based architectures and Large Language Models (LLMs) are pre-trained on large-scale datasets, the underlying model reasoning is not transparent to the legal expert. This paper jointly addresses court judgment prediction and explanation by not only predicting the judgment but also providing legal experts (...) with sentence-based explanations. To boost the performance of both tasks we leverage a legal named entity recognition step, which automatically annotates documents with meaningful domain-specific entity tags and masks the corresponding fine-grained descriptions. In such a way, transformer-based architectures and Large Language Models can attend to in-domain entity-related information in the inference process while neglecting irrelevant details. Furthermore, the explainer can boost the relevance of entity-enriched sentences while limiting the diffusion of potentially sensitive information. We also explore the use of in-context learning and lightweight fine-tuning to tailor LLMs to the legal language style and the downstream prediction and explanation tasks. The results obtained on a benchmark dataset from the Indian judicial system show the superior performance of entity-aware approaches to both judgment prediction and explanation. (shrink)

Condensing the key message conveyed by a long document into an informative summary is particularly helpful to lawyers and legal experts. State-of-the-art approaches to legal document summarization rely on Language Models (LMs) and are mostly trained on English documents. More limited research efforts have been devoted to summarizing legal documents in languages other than English. In this work, we investigate the applicability of Large Language Models (LLMs) to summarize Italian legal news documents. We benchmark state-of-the-art abstractive summarization techniques based on (...) Language Models, Large and not, for headline and abstract generation from legal news documents. We run an extensive set of experiments on a proprietary legal dataset, evaluating the resulting summaries according to both quantitative metrics and human evaluation. As expected, latest LLMs outperform classical models such as BART, T5, particularly in terms of grammaticality and informativeness of the summary content. Fine-tuned LLMs also show a significant increase in performance, variable across law areas, compared to their zero-shot setting. Importantly, the level of specialization of the fine-tuned version already reaches a steady state after feeding the model with few hundreds of training data. (shrink)

The ever-increasing volume of electronic legal documents calls for effective, language-specific summarization and headline generation techniques to make legal content more accessible and easy-to-use. In the context of Italian law existing summarization models are either extractive or focused on abstracting long-form summaries. As a result, the generated summaries have a low level of readability or are not suited to summarize common legal documents such as norms. This paper proposes LegItBART, a new abstractive summarization model. It leverages a BART-based sequence-to-sequence architecture (...) that is specifically pre-trained on Italian legal corpora. To enable the generation of concise summaries and headlines, we release two new annotated datasets tailored to the Italian legal domain, namely LawCodes and LegItConcepts. To successfully handle input documents exceeding the maximum token length such as verbose norms, codes, or legal articles, we also extend BART by integrating a global-sparse-local attention mechanism. We empirically analyze the performance of different pretraining and fine-tuned model combinations. The results show that using a mix of general-purpose and domain-specific pre-training yields relevant summarization performance improvements. The fine-tuned version of LegItBART outperforms all the tested baselines even those characterized by a significantly higher number of model parameters. (shrink)