Case law retrieval is the task of locating truly relevant legal cases given an input query case. Unlike information retrieval for general texts, this task is more complex with two phases (legal case retrieval and legal case entailment) and much harder due to a number of reasons. First, both the query and candidate cases are long documents consisting of several paragraphs. This makes it difficult to model with representation learning that usually has restriction on input length. Second, the concept of (...) relevancy in this domain is defined based on the legal relation that goes beyond the lexical or topical relevance. This is a real challenge because normal text matching will not work. Third, building a large and accurate legal case dataset requires a lot of effort and expertise. This is obviously an obstacle to creating enough data for training deep retrieval models. In this paper, we propose a novel approach called supporting model that can deal with both phases. The underlying idea is the case–case supporting relation and the paragraph–paragraph as well as the decision-paragraph matching strategy. In addition, we propose a method to automatically create a large weak-labeling dataset to overcome the lack of data. The experiments showed that our solution has achieved the state-of-the-art results for both case retrieval and case entailment phases. (shrink)

Motivated by the increasing economic significance of investment advisory industries and the prevalence of wrongdoing in financial planning services, we examine whether, and to what extent, employee misconduct is shaped by their local corruption culture. Using novel data of more than 4.7 million adviser-year observations of financial advisers and the Department of Justice’s data on corruption, we find that financial advisers and advisory firms located in areas with higher levels of corruption are more likely to commit misconduct. These results hold (...) for both individual advisor and firm level analyses and are robust to the use of various fixed effects, model specifications, proxies for corruption and misconduct, and an instrumental variable approach. Using the passage of the Dodd-Frank Whistleblower Provision, which provides incentives for reporting corruption incidences and thereby reduces the incentives for fraud, we find that the relation between local corruption culture and adviser misconduct is attenuated after the provision enacted by the SEC. Overall, our study highlights the externalities of corruption culture on individual ethics and the essential role of whistleblowing laws in reducing corruption-prone norms. (shrink)

Legal text retrieval serves as a key component in a wide range of legal text processing tasks such as legal question answering, legal case entailment, and statute law retrieval. The performance of legal text retrieval depends, to a large extent, on the representation of text, both query and legal documents. Based on good representations, a legal text retrieval model can effectively match the query to its relevant documents. Because legal documents often contain long articles and only some parts are relevant (...) to queries, it is quite a challenge for existing models to represent such documents. In this paper, we study the use of attentive neural network-based text representation for statute law document retrieval. We propose a general approach using deep neural networks with attention mechanisms. Based on it, we develop two hierarchical architectures with sparse attention to represent long sentences and articles, and we name them Attentive CNN and Paraformer. The methods are evaluated on datasets of different sizes and characteristics in English, Japanese, and Vietnamese. Experimental results show that: (i) Attentive neural methods substantially outperform non-neural methods in terms of retrieval performance across datasets and languages; (ii) Pretrained transformer-based models achieve better accuracy on small datasets at the cost of high computational complexity while lighter weight Attentive CNN achieves better accuracy on large datasets; and (iii) Our proposed Paraformer outperforms state-of-the-art methods on COLIEE dataset, achieving the highest recall and F2 scores in the top-N retrieval task. (shrink)

Background: Little is known about U.S. healthcare provider views and practices regarding evidence, counseling, and shared decision-making about in-hospital versus out-of-hospital birth settings. Methods: We conducted 19 in-depth, semistructured, qualitative interviews of eight obstetricians, eight midwives, and three pediatricians from across the United States. Interviews explored healthcare providers’ interpretation of the current evidence and their personal and professional experiences with childbirth within the existing medical, ethical, and legal context in the United States. Results: Themes emerged concerning risks and benefits, decision-making, (...) and patient-provider power dynamics. Collectively, the narratives illuminated fundamental ideological tensions between in- and out-of-hospital providers arising from divergent assignment of value to described risks and benefits. The majority of physicians focused on U.S.-specific data demonstrating increased neonatal morbidity and mortality associated with delayed access to hospital-based interventions, thereby justifying hospital birth as the standard of care. By contrast, midwives emphasized data demonstrating fewer interventions and superior maternal and neonatal outcomes in high-income European countries, where out-of-hospital birth is more common for low-risk birthing people. A key gap in counseling was revealed, as no interviewees offered anticipatory counseling regarding birth setting options. Providers directly and indirectly illustrated the propensity for asymmetric power relations between birth providers and pregnant people, especially in hospital settings. Conclusions: The narratives highlight the common goal of optimizing maternal and neonatal outcomes despite tensions arising from divergent prioritization of specific maternal and neonatal risks. Our findings suggest opportunities to foster collaboration and optimize outcomes via mutual respect and improved integration of care. (shrink)

Words not only define our reality, they reveal and direct our attitudes toward that reality. This feature of language is especially problematic when it comes to our treatment of animals. In this supple, thoughtful, and closely argued work, Vietnamese-American scholar Hanh Nguyen examines the words we use in English and other languages to justify our abuse of other-than-human world. Reframing our metaphors, categories, and descriptions of animals we love, hate, and eat allows us to break long-standing habits and see our (...) obligations and connections to non-human animals more clearly. (shrink)