Legal NLP

The importance of clear and correct text in legal documents cannot be understated, and, consequently, a grammatical error correction tool meant to assist a professional in the law must have the ability to understand the possible errors in the context of a legal environment, correcting them accordingly, and implicitly needs to be trained in the same environment, using realistic legal data. However, the manually annotated data required by such a process is in short supply for languages such as Romanian, much less for a niche domain. The most common approach is the synthetic generation of parallel data; however, it requires a structured understanding of the Romanian grammar. In this paper, we introduce, to our knowledge, the first Romanian-language parallel dataset for the detection and correction of grammatical errors in the legal domain, RoLegalGEC, which aggregates 350,000 examples of errors in legal passages, along with error annotations. Moreover, we evaluate several neural network models that transform the dataset into a valuable tool for both detecting and correcting grammatical errors, including knowledge-distillation Transformers, sequence tagging architectures for detection, and a variety of pre-trained text-to-text Transformer models for correction. We consider that the set of models, together with the novel RoLegalGEC dataset, will enrich the resource base for further research on Romanian.

We introduce LegalBench-BR, the first public benchmark for evaluating language models on Brazilian legal text classification. The dataset comprises 3,105 appellate proceedings from the Santa Catarina State Court (TJSC), collected via the DataJud API (CNJ) and annotated across five legal areas through LLM-assisted labeling with heuristic validation. On a class-balanced test set, BERTimbau-LoRA, updating only 0.3% of model parameters, achieves 87.6% accuracy and 0.87 macro-F1 (+22pp over Claude 3.5 Haiku, +28pp over GPT-4o mini). The gap is most striking on administrativo (administrative law): GPT-4o mini scores F1 = 0.00 and Claude 3.5 Haiku scores F1 = 0.08 on this class, while the fine-tuned model reaches F1 = 0.91. Both commercial LLMs exhibit a systematic bias toward civel (civil law), absorbing ambiguous classes rather than discriminating them, a failure mode that domain-adapted fine-tuning eliminates. These results demonstrate that general-purpose LLMs cannot substitute for domain-adapted models in Brazilian legal classification, even when the task is a simple 5-class problem, and that LoRA fine-tuning on a consumer GPU closes the gap at zero marginal inference cost. We release the full dataset, model, and pipeline to enable reproducible research in Portuguese legal NLP.

Despite significant advances in large language models, personalizing them for individual decision-makers remains an open problem. Here, we introduce a synthetic-organic supervision pipeline that transforms raw judicial decisions into instruction-tuning data, enabling parameter-efficient fine-tuning of personalized models for individual judges in low-resource settings. We compare our approach to state-of-the-art personalization techniques across three different tasks and settings. The results show that Causal Language Modeling followed by synthetically generated instruction-tuning significantly outperforms all other baselines, providing significant improvements across lexical, stylistic, and semantic similarity. Notably, our model-generated outputs are indistinguishable from the reasoning of human judges, highlighting the viability of efficient personalization, even in low-resource settings.

Legal domains in high-resource languages like English have widely adopted artificial intelligence for legal question answering. However, data scarcity in low resource languages such as Nepali has limited the training of large language models on Nepali legal texts. This study presents the first application of a Retrieval Augmented Generation based model for Nepali legal question answering using case laws extracted from the Nepal Kanun Patrika digital archive. Using BM25 on chunked documents, the approach achieved a top precision at one of 91 percent, and up to 75 percent with the multilingual E5 large model. Evaluation of generated answers showed 74 percent groundedness, 85 percent truthfulness according to an automated judge model, and 84 percent human evaluated truthfulness when using BM25 document retrieval, with a 92 percent successful answer generation rate. These results demonstrate that the RAG pipeline can effectively address the gap in legal question answering for low resource languages and provide a foundation for reliable AI systems in the Nepali legal domain.

Legal text generated by large language models (LLMs) can usually achieve reasonable factual accuracy, but it frequently fails to adhere to the specialised stylistic norms and linguistic conventions of legal writing. In order to improve stylistic quality, a crucial first step is to establish a reliable evaluation method. However, having legal experts manually develop such a metric is impractical, as the implicit stylistic requirements in legal writing practice are difficult to formalise into explicit rubrics. Meanwhile, existing automatic evaluation methods also fall short: reference-based metrics conflate semantic accuracy with stylistic fidelity, and LLM-as-a-judge evaluations suffer from opacity and inconsistency. To address these challenges, we introduce CLASE (Chinese LegAlese Stylistic Evaluation), a hybrid evaluation method that focuses on the stylistic performance of legal text. The method incorporates a hybrid scoring mechanism that combines 1) linguistic feature-based scores and 2) experience-guided LLM-as-a-judge scores. Both the feature coefficients and the LLM scoring experiences are learned from contrastive pairs of authentic legal documents and their LLM-restored counterparts. This hybrid design captures both surface-level features and implicit stylistic norms in a transparent, reference-free manner. Experiments on 200 Chinese legal documents show that CLASE achieves substantially higher alignment with human judgments than traditional metrics and pure LLM-as-a-judge methods. Beyond improved alignment, CLASE provides interpretable score breakdowns and suggestions for improvements, offering a scalable and practical solution for professional stylistic evaluation in legal text generation (Code and data for CLASE is available at: https://github.com/rexera/CLASE).

Large language models (LLMs) now support contexts of up to 1M tokens, but their strengths and weaknesses on complex long-context tasks remain unclear. To study this, we focus on multi-document legal case summarization, where a single case often spans many documents exceeding 100K tokens. We systematically evaluate 12 frontier LLMs with Gavel, which consists of Gavel-Ref, a reference-based evaluation framework with checklist, residual-fact, and writing-style evaluations, and Gavel-Agent, a reference-free agent for evaluating factual coverage directly from source documents. Our results show that current models are more prone to omitting key information than hallucinating. They all perform well on simple checklist items, such as filing date, but struggle with rare and complex items, such as settlements. Performance also declines as case length increases. To meta-evaluate Gavel, we collect 160 hours of human annotations. Gavel-Agent reduces token usage by at least 36% compared to end-to-end and chunk-by-chunk methods while achieving competitive performance. Gavel-Agent also generalizes to the medical domain, performing the best with at least 77% fewer tokens.

Statutory corpora and judicial decisions are growing faster than legal professionals can read them, while individual judgments often exceed the context limits of standard encoder models. Transformer architectures dominate legal NLP benchmarks, but their quadratic attention complexity can require truncating or fragmenting documents that demand whole-document reasoning. Selective state-space models (SSMs), such as Mamba, offer linear-time sequence modeling and are a promising alternative for long legal documents, yet their performance on legal classification and retrieval remains underexplored. We present a preliminary benchmark comparing Mamba and SSD-Mamba with BERT, DeBERTa, and Longformer across four legal classification tasks (ECtHR, EUR-Lex, SCOTUS, and ILDC/ILC) and two case-retrieval tasks (ECtHR and ILDC), using a shared windowing and aggregation pipeline. The strongest SSM performs within approximately 1.3 percentage points of the strongest transformer across tasks and metrics. SSD-Mamba achieves the best results on most metrics for ECtHR classification, ILDC classification, and ECtHR retrieval, while processing approximately 3 times more tokens per second than DeBERTa and 4 times more than Longformer. DeBERTa remains strongest on SCOTUS and EUR-Lex F1. These results are preliminary because they do not include variance estimates across random seeds or statistical significance testing. Rather than presenting a definitive ranking, we use these findings to motivate further evaluation with repeated-seed experiments, statistical testing, and controls for model capacity and computational efficiency.