https://repositorio.ufu.br/handle/123456789/5170 2026-07-16T01:10:34Z 2026-07-16T01:10:34Z https://repositorio.ufu.br/handle/123456789/48883 2026-07-15T20:59:03Z 2026-02-23T00:00:00Z

Title: Memória discursiva computacional: entre a consignação e o silenciamento do feminismo nas textualidades e visualidades algorítmicas em modelos de linguagem de larga escala Abstract: This thesis aims to propose, develop, and demonstrate the concept of computational discursive memory through an analysis of the linguistic and discursive functioning of large language models (LLMs), with particular attention to ChatGPT and Gemini. In a context in which these models have become central actors in the production of textual and imagistic utterances, it is urgent to investigate how they operate, non-neutrally, upon discursive archives, and to what extent they reproduce, silence, or divert meanings. The study begins from the hypothesis that the language generated by artificial intelligence systems, although grounded in mathematical–statistical techniques, configures a form of discursive memory that is analogous, though not identical, to that defined within the tradition of materialist Discourse Analysis (Pêcheux, 1995, 1997, 2006). This memory, rather than merely storing data, functions as a condition of possibility for saying, enabling the emergence of articulated effects among historically inscribed utterances. By examining whether it is possible to speak of a machine memory that operates discursively, the thesis offers an innovative approach at the intersection of linguistics, computer science, and critical technology studies. The increasing adoption of artificial intelligence systems across diverse social domains has naturalized the idea that machines “understand” and “respond” linguistically. Yet this operation is grounded in massive volumes of linguistic data that bear historically constituted discursive formations and function as archives. In this sense, artificial intelligence does not “create” in a vacuum but draws upon what has already been said. The research problem, then, lies in investigating and analyzing: can memory in LLMs be conceptualized as discursive? If so, how this concept contributes to understanding artificial language as a symbolic and ideological process. The justification is grounded in the recognition that these models operate on pre-existing, socially situated utterances, systematically recombining sequences of words that reactivate historically sedimented meanings and discursive positions. By challenging the traditional conception of memory in computer science, restricted to technical storage and retrieval, the thesis proposes a critical, political, and epistemic expansion of the concept. Furthermore, as a researcher, woman, situated within a research group dedicated to body, language, and technologies, the choice of feminism as a thematic and analytical axis for testing the textual and imagistic meaning effects generated by these LLMs is justified both epistemologically and through personal and political implication. From this enunciative position, the study seeks to show how the erasure of certain forms of existence and resistance in artificial intelligence generated outputs exposes the limits, and dangers, of computational discursive memory. The research adopts a qualitative, critical, and interpretivist approach, articulating discourse analysis with experimentation in natural language and image generation. The study draws on LLMs available on public platforms, using prompts designed to challenge their ability to generate ideologically marked meanings. The corpus comprises texts and images generated from these prompts, whose analyzed data reveal consistent patterns of reproducing hegemonic discourses, erasing antagonisms, and producing symbolic sanitization. This suggests that generative artificial intelligence relies on utterances aligned with dominant discourses present in training data, thereby avoiding dissent. In the domain of image generation, the models consistently produced white, young, slim, Westernized bodies. The confrontation between these outputs and the theory of discursive memory reveals that the architecture of LLMs favors the repetition of historically dominant meanings: by operating on statistical patterns, the models produce ideological stabilization effects, reinforcing naturalized discourses without intention or interpretation. The notion of computational discursive memory therefore does not operate as a metaphor, but as an analytical concept that makes it possible to critically examine how statistical regularities resulting from the model architectures generate effects that can be reinscribed within the discursive field. When such patterns are put into circulation in dialogue with human subjects, they acquire symbolic density, for they are appropriated, challenged, displaced, and reinscribed in discursive practices that produce meaning effects, including those of subjectivation, erasure, or resistance. The theoretical contribution of the thesis lies in updating the concept of discursive memory for a new empirical domain, namely, algorithmic language technologies. Its political contribution lies in exposing the erasures and neutralizations these technologies enact, often under the guise of technical neutrality. Its methodological contribution lies in integrating discourse analysis, technology studies, and critical experimentation with artificial intelligence systems.

2026-02-23T00:00:00Z https://repositorio.ufu.br/handle/123456789/48876 2026-07-15T17:12:46Z 2026-02-27T00:00:00Z

Title: Mediação no processo de ensino e aprendizagem de língua portuguesa para surdos: a produção de texto dissertativo-argumentativo para o Enem Abstract: Since 2017, there has been a version of the National High School Exam (ENEM) adapted for deaf candidates who use Brazilian Sign Language (Libras) as their first language (L1). Despite this, there is still a low presence of deaf students at the Federal University of Triângulo Mineiro (UFTM). This research, developed from an extension course offered by UFTM entitled “Text Production for Deaf Students,” aims to understand the teaching and learning process of deaf participants in the production of argumentative texts, according to the model required by ENEM, in Portuguese, through a proposal developed on the Moodle platform. The specific objectives were to identify relevant aspects of the structure of an argumentative text present in the written productions of deaf participants in the course; to discuss the learning process of argumentative writing in light of the challenges presented by the deaf participants; and to analyze the participants’ perceptions regarding the learning of argumentative text production. Adopting a qualitative and interpretivist approach, this study employed the following data collection instruments: an initial interview conducted before the course, written production activities, a diary, and a final interview conducted after the end of the course. The participants were deaf students who had completed or were currently attending high school and who fully completed the proposed activities. The data revealed that most participants had late contact with Libras, limited mediation by interpreters, and reduced reading and writing practices throughout their schooling. These factors affected their learning of Portuguese as a second language and generated insecurity in textual production. Analyses of the written productions indicated recurring challenges in the structural organization of the text, in the formulation of the thesis statement, in the articulation of arguments, and in the inclusion of evidence, as well as interlanguage features resulting from the influence of Libras. These results indicate that the development of writing skills requires extended time for practice, linguistic mediation in Libras, continuous feedback, and opportunities for rewriting. On the other hand, interaction between participants and teachers in Libras, the use of multimodal resources, visual materials, and synchronous and asynchronous moments in the virtual environment promoted greater participation, improved comprehension of the content, and gradual progress in textual production. Such progress was observed through the expansion of information, the inclusion of titles, the use of concrete data, and better textual organization. The results reinforce the importance of bilingual pedagogical practices, continuous teacher education, and the central role of Libras as a language of instruction, communication, and knowledge mediation. This study contributes to the field of Applied Linguistics by offering reflections on the teaching of Portuguese to deaf students, considering accessibility through Libras and the use of virtual learning platforms, the mediating role of the teacher, and the relevance of the educational trajectory of deaf students.

2026-02-27T00:00:00Z https://repositorio.ufu.br/handle/123456789/48873 2026-07-15T15:55:14Z 2026-02-24T00:00:00Z

Title: Paulo Leminski como sintoma de seu tempo Abstract: This dissertation examines the work of Paulo Leminski based on the hypothesis that his poetics are structured through aesthetic and discursive tensions, functioning as a symptom of the cultural, political, and literary transformations that marked Brazil from the 1960s to the 1980s. Frequently categorized as a concrete, marginal, pop, or postmodern poet, Leminski is here understood as an author who occupies in-between spaces, articulating formal rigor and colloquial language, verbo-visual experimentation and orality, modern desubjectivation and the fragmented return of the lyrical self. The study adopts an analytical focus on selected passages from works such as Catatau, Quarenta Clics em Curitiba, Caprichos e Relaxos, Distraídos Venceremos, and Ensaios e Anseios Crípticos, highlighting moments in which these tensions become formally explicit. The theoretical framework draws on discussions of concrete poetry, aesthetics, form, and subjectivity, engaging with authors such as Haroldo and Augusto de Campos, Décio Pignatari, Michel Collot, Marcos Siscar, and Vladimir Safatle. The dissertation argues that the contradictions within Leminski’s work do not undermine its coherence; rather, they constitute its creative logic, transforming conflict into method. In doing so, it demonstrates that Leminski not only reflects his time but formally stages it, producing a poetics that remains essential to understanding contemporary Brazilian literature.

2026-02-24T00:00:00Z https://repositorio.ufu.br/handle/123456789/48791 2026-06-30T06:18:22Z 2025-12-05T00:00:00Z

Title: FicMed - Vocabulário do drama médico na Tv: um estudo baseado em corpus Abstract: This dissertation examines the constitution, organization and diachronic variation of the medical-fictional vocabulary found in English-language television medical dramas produced between the 1980s and the 2020s, based on the FicMed corpus and the terminographic platform VoTec. Grounded in the Communicative Theory of Terminology (CTT) and Corpus Linguistics, the study analyzes how medical terminology is appropriated and recontextualized within fictional narratives, highlighting the relationship between language, science and culture. The findings indicate that the medical-fictional lexicon is structured around a relatively stable core of terms, such as Aids, Blood Pressure, CT Scan, Leukemia and Lupus, alongside a set of emerging terms that reflect scientific, social and discursive transformations across decades. The diachronic analysis reveals semantic-discursive shifts and demonstrates how medical discourse is translated into the popular imagination, contributing to the social circulation and humanization of specialized language. It is argued that medical television series operate as mediators of scientific knowledge by transforming technical terminology into narrative and symbolic elements. The study expands the field of Terminology by situating specialized language at the intersection of discourse, culture and narrative, while also proposing a methodological model for the terminographic analysis of multimodal corpora.

2025-12-05T00:00:00Z