Retrieving case law is a time-consuming task predominantly carried out by querying databases. We provide a comparison of two models in three different settings for Czech Constitutional Court decisions: (i) a large general-purpose embedder (OpenAI), (ii) a domain-specific BERT-trained from scratch on ~30,000 decisions using sliding windows and attention pooling. We propose a noise-aware evaluation including IDF-weighted keyword overlap as graded relevance, binarization via two thresholds (0.20 balanced, 0.28 strict), significance via paired bootstrap, and an nDCG diagnosis supported with qualitative analysis. Despite modest absolute nDCG (expected under noisy labels), the general OpenAI embedder decisively outperforms the domain pre-trained BERT in both settings at @10/@20/@100 across both thresholds; differences are statistically significant. Diagnostics attribute low absolutes to label drift and strong ideals rather than lack of utility. Additionally, our framework is robust enough to be used for evaluation under a noisy gold dataset, which is typical when handling data with heterogeneous labels stemming from legacy judicial databases.

The growing intersection of cybersecurity and law creates a complex information space where traditional legal research tools struggle to deal with nuanced connections between cases, statutes, and technical vulnerabilities. This knowledge divide hinders collaboration between legal experts and cybersecurity professionals. To address this important gap, this work provides a first step towards intelligent systems capable of navigating the increasingly intricate cyber-legal domain. We demonstrate promising initial results on multilingual tasks.

Machine learning is increasingly used in the legal domain, where it typically operates retrospectively by treating past case outcomes as ground truth. However, legal outcomes are often shaped by human interventions that are not captured in most machine learning approaches. A final decision may result from a settlement, an appeal, or other procedural actions. This creates label indeterminacy: the outcome could have been different if the intervention had or had not taken place. We argue that legal machine learning applications need to account for label indeterminacy. Methods exist that can impute these indeterminate labels, but they are all grounded in unverifiable assumptions. In the context of classifying cases from the European Court of Human Rights, we show that the way that labels are constructed during training can significantly affect model behaviour. We therefore position label indeterminacy as a relevant concern in AI & Law and demonstrate how it can shape model behaviour.

Legal AI systems are increasingly being adopted by judicial and legal system deployers and providers worldwide to support a range of applications. While they offer potential benefits such as reducing bias, increasing efficiency, and improving accountability, they also pose significant risks, requiring a careful balance between opportunities, and legal and ethical development and deployment. AI literacy, as a legal requirement under the EU AI Act and a critical enabler of ethical AI for deployers and providers, could be a tool to achieve this. The article introduces the term "legal AI systems" and then analyzes the concept of AI literacy and the benefits and risks associated with these systems. This analysis is linked to a broader AI-L concept for organizations that deal with legal AI systems. The outcome of the article, a roadmap questionnaire as a practical tool for developers and providers to assess risks, benefits, and stakeholder concerns, could be useful in meeting societal and regulatory expectations for legal AI.

This report addresses the challenge of limited labeled datasets for developing legal recommender systems, particularly in specialized domains like labor disputes. We propose a new approach leveraging the co-citation of legal articles within cases to establish similarity and enable algorithmic annotation. This method draws a parallel to the concept of case co-citation, utilizing cited articles as indicators of shared legal issues. To evaluate the labeled results, we employ a system that recommends similar cases based on plaintiffs' accusations, defendants' rebuttals, and points of disputes. The evaluation demonstrates that the recommender, with finetuned text embedding models and a reasonable BiLSTM module can recommend labor cases whose similarity was measured by the co-citation of the legal articles. This research contributes to the development of automated annotation techniques for legal documents, particularly in areas with limited access to comprehensive legal databases.

There is a surge observed in research and applications of computer science and artificial intelligence in the legal domain. The related term computational law is commonly defined as "the branch of Legal Informatics concerned with the representation of rule and regulations in computable form" [Genesereth and Chaudhri, 2022]. The focus of an important percentage of related work on computational law is on automatic processing, generation, or understanding of legal documents [Küçük and Can, 2024]. Recent advancements in artificial intelligence (AI), such as generative AI models, pre-trained language models (PLMs) or large language models (LLMs), and chatbots developed using such models, have also affected the domain of computational law, and this dramatic impact is also acknowledged by legal professionals [Goth, 2024]. Undoubtedly, annotated or unannotated datasets and benchmarks in digital form are required for legal AI studies on legal texts, in order to facilitate model training, and to ensure sound comparisons of different approaches to the problems pertaining to computational law.

This paper aims to offer AI & Law researchers and practitioners a more detailed understanding of whether and how continued pre - training and instruction fine - tuning (IFT) of large language models (LLMs) on legal corpora increases their utilization of human - defined legal concepts when developing global contextual representations of input sequences. We compare d three models: Mistral 7B, SaulLM - 7B - Base (Mistral 7B with continued pre - training on legal corpora), and SaulLM - 7B - Instruct (with further IFT). T his preliminary assessment examine d 7 distinct text sequences from recent AI & Law literature, each containing a human - defined legal concept. We first compared the proportions of total attention the models allocated to subsets of tokens representing the legal concepts. We then visualized patterns of raw attention score alterations, evaluating whether legal training introduce d novel attention patterns corresponding to structures of human legal knowledge. This inqu i ry revealed that (1) the impact of legal training was unevenly distributed across the various human - defined legal concepts, and (2) the contextual representations of legal knowledge learned during legal training did not coincide with structures of human - defined legal concepts. We conclude with suggestions for further investigation into the dynamics of legal LLM training .

Factors are a foundational component of legal analysis and computational models of legal reasoning. These factor-based representations enable lawyers, judges, and AI and Law researchers to reason about legal cases. In this paper, we introduce a methodology that leverages large language models (LLMs) to discover lists of factors that effectively represent a legal domain. Our method takes as input raw court opinions and produces a set of factors and associated definitions. We demonstrate that a semi-automated approach, incorporating minimal human involvement, produces factor representations that can predict case outcomes with moderate success, if not yet as well as expert-defined factors can.

The concept of code-driven law, i.e. of "legal norms or policies that have been articulated in computer code" by some actors with normative competence, has been convincingly elaborated by Hildebrandt [1]. Its introduction has the merit to refocus the discussion on the role of artificial devices in the legal activity, rather than on ontological positions expressed under code-is-law or law-is-code banners, which are present, with various interpretations and changing fortunes, in the literature and practice of contemporary regulatory technologies, and technology-oriented legal scholarship (see the overview in [2]). According to Hildebrandt, code-driven law should be distinguished from data-driven law, i.e. computational decision-making derived from statistical or other inductive methods, and from text-driven law, i.e. the legal activity performed by humans by means of sources of norms such as statutory and case law. A crucial difference between these forms of "law" is that the linguistic artifacts used in text-driven law are characterized by open-textured concepts (e.g.

Legal intake, the process of finding out if an applicant is eligible for help from a free legal aid program, takes significant time and resources. In part this is because eligibility criteria are nuanced, open-textured, and require frequent revision as grants start and end. In this paper, we investigate the use of large language models (LLMs) to reduce this burden. We describe a digital intake platform that combines logical rules with LLMs to offer eligibility recommendations, and we evaluate the ability of 8 different LLMs to perform this task. We find promising results for this approach to help close the access to justice gap, with the best model reaching an F1 score of .82, while minimizing false negatives.