{"id":228,"date":"2025-03-12T08:49:23","date_gmt":"2025-03-12T08:49:23","guid":{"rendered":"https:\/\/cilc2025.usal.es\/?page_id=228"},"modified":"2025-05-02T07:26:21","modified_gmt":"2025-05-02T07:26:21","slug":"plenarios","status":"publish","type":"page","link":"https:\/\/cilc2025.usal.es\/en\/plenarios\/","title":{"rendered":"Keynote Speakers"},"content":{"rendered":"\n
<\/i><\/i>Marc Alexander and James Balfour<\/h5><\/a><\/div>
\n

 <\/p>\n

\"\"<\/p>\n

\u2018up yours at the end of life\u2019: Opposition, Emotion, and Overlap in a Corpus of Scottish Debates on\u00a0Assisted\u00a0Dying<\/strong><\/span><\/p>\n

Marc Alexander and James Balfour<\/span>
\nUniversity of Glasgow<\/span><\/p>\n

In this plenary, we report on findings from a recent corpus-based study examining public discourse surrounding the Assisted Dying for Terminally Ill Adults (Scotland) Bill, which proposes to legalize medically assisted death for mentally competent terminal patients. With public opinion sharply divided\u2014faith groups generally opposing while public polls show over 70% support\u2014the study analyzes both public consultation responses and media coverage to understand key attitudes and arguments.<\/span><\/p>\n

The research comprises two complementary studies. The first analyzes public responses to government consultations, drawing from two datasets: 12,314 written submissions from 2022 (2.1 million words) and 7,236 responses from 2024 (1.8 million words). The second examines the media narrative around assisted dying between 2022-2024 in the UK (6,360 texts). By examining language patterns in the two datasets in tandem we reflect on the complex interaction between public attitudes towards a sensitive and contentious topic and the role media framing plays in shaping public debate.<\/span><\/p>\n

To examine oppositional discourse in the public responses, we first identify unique lexical choices exclusive to each group\u2014supporters used terms like “hideous,” “abject,” “urine,” and “linger,” while opponents employed words such as “eroded,” “burden,” “wedge,” and “shalt.” Second, tagging the corpus using WMatrix, we compare key semantic domains between groups, revealing that supporters’ responses contained more emotional content (particularly in the “Sad” domain), while opponents’ responses were more analytical in nature. Third, we conduct n-gram studies to identify areas of common ground between opposing viewpoints and detect potential copy-pasted responses within each group.<\/span><\/p>\n

For the second part of the study, we investigate media influence on the debate by analyzing 6,360 news articles published between 2022-2024 that explicitly referenced assisted dying. Using keyword analysis, n-grams, and concordance analysis, we examine how different politically-affiliated newspapers framed the debate. Our findings suggest significant overlap between media narratives and public consultation responses, with some phraseology being nearly identical across both datasets.<\/span><\/p>\n

The research extends beyond the specific assisted dying debate to address broader methodological questions about analyzing oppositional discourse in public life. The study revealed distinct rhetorical patterns: supporters of the bill tended to employ more emotionally charged language and personal narratives, while opponents favored analytical and consequence-based arguments. The media analysis demonstrated how news coverage might reinforce these polarized positions through consistent framing patterns. The bill’s consideration in Scotland occurs against the backdrop of similar debates in England, where recent legislative efforts have faced setbacks. The research notes the particular challenges of analyzing representative corpora in highly polarized debates, where responses may range from deeply personal experiences to organized campaign submissions.<\/span><\/p>\n

This comprehensive analysis of both public and media discourse provides valuable insights into how contentious healthcare policy debates are framed and argued across different platforms and stakeholder groups. The methodological approach developed here offers a framework for analyzing other polarized public debates, while the findings contribute to our understanding of how public opinion forms and expresses itself on complex ethical issues.<\/span><\/p>\n<\/div><\/div><\/div>\n

<\/i><\/i>Pascual Cantos<\/h5><\/a><\/div>
\n

\"\"<\/p>\n

La inteligencia artificial generativa en el laberinto de la traducci\u00f3n especializada<\/strong><\/span><\/p>\n

Pascual Cantos<\/span>
\nUniversidad de Murcia<\/span><\/p>\n

En este estudio se analiza la eficacia de la inteligencia artificial (IA) generativa como herramienta de traducci\u00f3n autom\u00e1tica aplicada a dominios especializados. La investigaci\u00f3n se fundamenta en un enfoque metodol\u00f3gico que combina la ling\u00fc\u00edstica de corpus con t\u00e9cnicas cuantitativas avanzadas, empleando corpus paralelos en los campos biom\u00e9dico, jur\u00eddico y t\u00e9cnico.\u00a0Estos corpus permiten evaluar<\/strong>\u00a0la capacidad de los sistemas de IA generativa para abordar tareas complejas como la precisi\u00f3n l\u00e9xica, la cohesi\u00f3n discursiva y la adaptaci\u00f3n contextual, aspectos cr\u00edticos en la traducci\u00f3n especializada.<\/span><\/p>\n

El dise\u00f1o de la investigaci\u00f3n se centra en la implementaci\u00f3n de an\u00e1lisis cuantitativos exhaustivos. Entre las t\u00e9cnicas empleadas destaca el an\u00e1lisis de frecuencia l\u00e9xica y de patrones de n-gramas, ambos dirigidos a identificar la consistencia terminol\u00f3gica y\u00a0a detectar posibles incoherencias<\/strong>\u00a0en el uso del vocabulario t\u00e9cnico. Para la evaluaci\u00f3n de la calidad de las traducciones, se aplican m\u00e9tricas ampliamente reconocidas en el campo de la traducci\u00f3n autom\u00e1tica, como BLEU (Bilingual Evaluation Understudy<\/em>) y COMET (Crosslingual Optimized Metric for Evaluation of Translation<\/em>). BLEU eval\u00faa la correspondencia entre las traducciones generadas y\u00a0las<\/strong>\u00a0traducciones de referencia, mientras que COMET incorpora un enfoque m\u00e1s matizado basado en modelos neuronales que predicen la calidad de la traducci\u00f3n tomando como referencia juicios humanos.<\/span><\/p>\n

Asimismo, el estudio adopta un an\u00e1lisis de varianza para explorar la adecuaci\u00f3n contextual de las traducciones generadas en los diferentes dominios. Este enfoque permite medir c\u00f3mo los modelos de IA gestionan variaciones en el contexto ling\u00fc\u00edstico y aseguran la coherencia discursiva, un requisito clave en los textos especializados. La metodolog\u00eda incluye, adem\u00e1s, una comparaci\u00f3n entre los resultados obtenidos por los sistemas de IA y las traducciones humanas, considerando par\u00e1metros como la precisi\u00f3n sem\u00e1ntica y la adaptabilidad estil\u00edstica.<\/span><\/p>\n

Un aspecto esencial del dise\u00f1o metodol\u00f3gico es la preparaci\u00f3n y curaci\u00f3n de los corpus empleados. Estos corpus paralelos se construyen a partir de fuentes autorizadas en los tres dominios seleccionados, asegurando una representaci\u00f3n adecuada de la terminolog\u00eda y los estilos discursivos espec\u00edficos de cada \u00e1rea. Se prioriza la inclusi\u00f3n de textos aut\u00e9nticos que reflejen un uso realista del lenguaje t\u00e9cnico, lo que facilita una evaluaci\u00f3n m\u00e1s precisa de las capacidades de los modelos generativos.\u00a0La elecci\u00f3n cuidadosa de estos textos es clave para garantizar la validez de las conclusiones extra\u00eddas.<\/strong><\/span><\/p>\n

El objetivo principal de este estudio es proporcionar una evaluaci\u00f3n integral de la IA generativa en la traducci\u00f3n autom\u00e1tica especializada. La investigaci\u00f3n no solo examina las capacidades actuales de estos modelos, sino que tambi\u00e9n sienta las bases para optimizar su rendimiento mediante la incorporaci\u00f3n de corpus de entrenamiento m\u00e1s espec\u00edficos y estrategias de evaluaci\u00f3n mejoradas.\u00a0Adem\u00e1s, busca establecer un marco metodol\u00f3gico que pueda ser replicado en estudios futuros sobre traducci\u00f3n autom\u00e1tica en otros dominios especializados.<\/strong>\u00a0Esta aproximaci\u00f3n metodol\u00f3gica, que combina an\u00e1lisis ling\u00fc\u00edsticos detallados con t\u00e9cnicas de evaluaci\u00f3n cuantitativa, contribuye a un entendimiento m\u00e1s profundo del potencial de la IA generativa en entornos de traducci\u00f3n profesional.<\/span><\/p>\n

<\/div><\/div><\/div>\n

<\/i><\/i>Mark Davies<\/h5><\/a><\/div>
\n

\"\"<\/p>\n

The Relevance of Large, Structured Corpora in the Age of Large Language Models<\/strong><\/span><\/p>\n

Mark Davies<\/span>
\nBrigham Young University<\/span><\/p>\n

I will provide a summary of the in-depth data from several \u201cwhite papers\u201d at English-Corpora.org,<\/u> on how well the predictions of two prominent Large Language Models (LLMs) match the actual data from several robust corpora, including corpora from Sketch Engine, and several corpora from English-Corpora.org (COCA, COCA, GloWbE, NOW, iWeb, the TV and Movie corpora, and more). I will also provide limited data from the three corpora in the Corpus del espa\u00f1ol and the three corpora in the Corpus do portugu\u00eas.<\/span><\/p>\n

In terms of strengths, the LLMs arguably provide:<\/span><\/p>\n