GUILHERME DA FRANCA COUTO FERNANDES DE ALMEIDA

GUILHERME DA FRANCA COUTO FERNANDES DE ALMEIDA

Lattes

ORCID

Scopus

Página do item completo

Resultados da Pesquisa

Agora exibindo 1 - 10 de 17

Uma Estratégia de Treinamento de Diretores Baseada em Dados
(2022) Ragazzo, Carlos Emmanuel Joppert; Guilherme da Franca Couto Fernandes de Almeida; GUILHERME DA FRANCA COUTO FERNANDES DE ALMEIDA
Neste artigo, descreve-se uma proposta para enfrentar o problema de equidade dentro de grandes redes municipais de ensino. Acredita-se que um programa de treinamento de diretores que use uma estratégia de mentoria entre pares seja capaz de melhorar o desempenho das escolas que apresentam consistentemente os piores resultados. Para defender a proposta, discuti-se experiências internacionais, trabalhos recentes que identificam pares de escolas geograficamente próximas, mas com desempenhos discrepantes e reporta-se os resultados de análises empíricas inéditas que revelam diferenças sistemáticas entre os diretores das escolas com performance boa e os diretores de escolas em risco.
A dual character theory of law
(2024) GUILHERME DA FRANCA COUTO FERNANDES DE ALMEIDA
One persistent question in jurisprudence relates to the role of morality in the concept of law. For instance, consider the question of whether unjust statutes are laws. Legal positivists say that they’re laws in every relevant sense, while natural lawyers say that they’re not. This article considers a different answer inspired by recent findings in experimental philosophy: there is one relevant sense in which unjust statutes are laws, but also a different relevant sense in which they aren’t. After considering the ways in which this alternative differs from some of the mainstream theories in general jurisprudence, the article argues that it provides more elegant solutions to two problems that have puzzled legal philosophers in the past: the paradox of customary international law and the shifts in legal discourse over history.
Do Formalist Judges Abide By Their Abstract Principles? A Two-Country Study in Adjudication
(2021) Bystranowski, Piotr; Janik, Bartosz; Próchnicki, Maciej; Hannikainen, Ivar Rodriguez; GUILHERME DA FRANCA COUTO FERNANDES DE ALMEIDA; Struchiner, Noel
Recent literature in experimental philosophy has postulated the existence of the abstract/concrete paradox (ACP): the tendency to activate inconsistent intuitions (and generate inconsistent judgment) depending on whether a problem to be analyzed is framed in abstract terms or is described as a concrete case. One recent study supports the thesis that this effect influences judicial decision-making, including decision-making by professional judges, in areas such as interpretation of constitutional principles and application of clear-cut rules. Here, following the existing literature in legal theory, we argue that the susceptibility to such an effect might depend on whether decision-makers operate in a legal system characterized by the formalist or particularist approach to legal interpretation, with formalist systems being less susceptible to the effect. To test this hypothesis, we compare the results of experimental studies on ACP run on samples from two countries differing in legal culture: Poland and Brazil. The lack of significant differences between those results (also for professional legal decision-makers) suggests that ACP is a robust effect in the legal context.
Exploring the psychology of LLMs’ moral and legal reasoning
(2024) GUILHERME DA FRANCA COUTO FERNANDES DE ALMEIDA; Nunes, José Luiz; Engelmann, Neele; Wiegmann, Alex; Araújo, Marcelo de
Large language models (LLMs) exhibit expert-level performance in tasks across a wide range of different domains. Ethical issues raised by LLMs and the need to align future versions makes it important to know how state of the art models reason about moral and legal issues. In this paper, we employ the methods of experimental psychology to probe into this question. We replicate eight studies from the experimental literature with instances of Google's Gemini Pro, Anthropic's Claude 2.1, OpenAI's GPT-4, and Meta's Llama 2 Chat 70b. We find that alignment with human responses shifts from one experiment to another, and that models differ amongst themselves as to their overall alignment, with GPT-4 taking a clear lead over all other models we tested. Nonetheless, even when LLM-generated responses are highly correlated to human responses, there are still systematic differences, with a tendency for models to exaggerate effects that are present among humans, in part by reducing variance. This recommends caution with regards to proposals of replacing human participants with current state-of-the-art LLMs in psychological research and highlights the need for further research about the distinctive aspects of machine psychology
Primeira Infância no Poder Judiciário
(2023) Berriel, Cecilia Machado; IVAR ALBERTO GLASHERSTER MARTINS LANGE HARTMANN; Marpin, Ábia; GUILHERME DA FRANCA COUTO FERNANDES DE ALMEIDA; Abbas, Lorena; Maia, Natália de Oliveira; Fabris, Ligia
Reconhecendo a importância do período de 0 a 6 anos de idade, o Marco Legal da Primeira Infância (MLPI) trata dos princípios e diretrizes para a elaboração de políticas públicas específicas para essa fase do desenvolvimento humano. Também estabelece alterações relevantes em outras normas já existentes, como o Estatuto da Criança e do Adolescente (ECA), a Consolidação das Leis do Trabalho (CLT) e o Código de Processo Penal (CPP), com vistas a privilegiar o melhor interesse da criança. Este livro reúne os resultados de uma pesquisa sobre a aplicação do MLPI em tribunais brasileiros que buscou responder, em uma perspectiva quantitativa: qual foi o impacto dessa norma sobre as decisões judiciais? E, em uma perspectiva qualitativa, quais as percepções de atores que trabalham no sistema de justiça sobre essa lei e suas possíveis implicações para a prática judicial.
O Direito sem respeito às regras
(2024) GUILHERME DA FRANCA COUTO FERNANDES DE ALMEIDA; Leite, Fábio Carvalho; Assis, Rodolfo
Neste artigo, apresentam-se os resultados de duas pesquisas quantitativas observacionais realizadas no Tribunal de Justiça do Rio de Janeiro. Com elas, investigaram-se processos envolvendo casos regulados pelo art. 20 do Código civil e pelo art. 19 do Marco civil da internet. Os resultados revelam que os juízes quase não mencionam esses dispositivos, mesmo em situações nas quais eles claramente se aplicam. Argumenta-se que esses resultados sinalizam uma atitude generalizada de desrespeito às regras primárias, o que torna o Direito brasileiro atípico do ponto de vista conceitual. Por fim, defende-se que, sob a perspectiva normativa, essa atipicidade é indesejável.
Epistemology Goes AI: A Study of GPT-3?s Capacity to Generate Consistent and Coherent Ordered Sets of Propositions on a Single-Input-Multiple-Outputs Basis
(2024) Araújo, Marcelo de; GUILHERME DA FRANCA COUTO FERNANDES DE ALMEIDA; Nunes, José Luiz
The more we rely on digital assistants, online search engines, and AI systems to revise our system of beliefs and increase our body of knowledge, the less we are able to resort to some independent criterion, unrelated to further digital tools, in order to asses the epistemic reliability of the outputs delivered by them. This raises some important questions to epistemology in general and pressing questions to applied to epistemology in particular. In this paper, we propose an experimental method for the assessment of GPT-3’s capacity to generate consistent and coherent sets of outputs. When several outputs to one and the same input are very repetitive they tend to be consistent with each other, that is they do not contradict each other. But consistency does not make the set of outputs as a whole more informative than the outputs considered individually. We argue that the less informative a set of outputs is, the less coherent it is. We establish a conceptual distinction between consistency and coherence in the light of what some epistemologists refer to as a coherence theories of truth and justification. While much attention has been given to GPT-3’s capacity to produce internally coherent individual outputs, we argue, instead, that more attention should be given to its capacity to produce consistent and coherent outputs generated on a single-input-multiple-outputs basis.
Extracting value from Brazilian Court decisions
(2022) Fernandes, William Paulo Ducca; Frajhof, Isabella Zalcberg; GUILHERME DA FRANCA COUTO FERNANDES DE ALMEIDA; Rodrigues, Ariane Moraes Bueno; Barbosa, Simone Diniz Junqueira; Konder, Carlos Nelson; Nasser, Rafael Barbosa; Carvalho, Gustavo Robichez de; Lopes, Hélio Côrtes Vieira
We propose a methodology to extract value from Brazilian Court decisions to support judges and lawyers in their decision-making. We instantiate our methodology in one information system we have developed. Such system (i) extracts plaintiff’s legal claims and each specific provision on legal opinions enacted by lower and Appellate Courts, and (ii) connects each legal claim with the corresponding judicial provision. The information system presents the results through visualizations. Information Extraction for legal texts has been previously approached in the literature for different languages, using different methods. Our proposal is different from previous work, since our corpora comprise Brazilian lower and Appellate Court decisions, in which we look for a set of plaintiff’s legal claims and judicial provisions commonly judged by the Court. We use the following methods to tackle the information extraction tasks: Bidirectional Long Short-Term Memory network; Conditional Random Fields; and a combination of Bidirectional Long Short-Term Memory network and Conditional Random Fields. In addition to the well-known distributed representation of words in word embeddings, we use character-level representation of words in character embeddings. We have built three corpora – Kauane Insurance Report, Kauane Insurance Lower, and Kauane Insurance Upper – to train and evaluate the system, using public data from the State Court of Rio de Janeiro. Our methods achieved good quality for Kauane Insurance Lower and Kauane Insurance Upper, and promising results for Kauane Insurance Report.
The Experimental Jurisprudence of the Concept of Rule: Implications for the Hart-Fuller Debate
(2023) GUILHERME DA FRANCA COUTO FERNANDES DE ALMEIDA; Struchiner, Noel; Hannikainen, Ivar R.
Demandas repetitivas
(2019) IVAR ALBERTO GLASHERSTER MARTINS LANGE HARTMANN; Pinheiro Júnior, Fausto Marques; GUILHERME DA FRANCA COUTO FERNANDES DE ALMEIDA; Araújo, Felipe; Correia Júnior, Fernando; Silva, Abner da