Portrait de Ayla Rigouts Terryn

Ayla Rigouts Terryn

Membre académique associé
Professeure adjointe, Université de Montréal, Linguistique et de traduction
Sujets de recherche
Évaluation linguistique des modèles de langage
Grands modèles de langage (LLM)
Recherche d'information
Terminologie
Traduction automatique
Traitement du langage naturel
Site web
Google Scholar
LinkedIn

Biographie

Ayla Rigouts Terryn est professeure adjointe en technologies de la traduction et en IA au département de linguistique et de traduction de l'Université de Montréal et membre académique associée à Mila. Elle est également professeure à IVADO (regroupement 3 : traitement du langage naturel (NLP)) et titulaire de la chaire IVADO-FRQ « Au carrefour des langues et de l'IA : vers une synergie entre l'expertise linguistique et l'innovation computationnelle ».

Elle a obtenu une maîtrise en traduction à l'université d'Anvers, un doctorat sur l'extraction automatique de terminologie à l'université de Gand et s'est spécialisée dans les technologies linguistiques multilingues en tant que chercheuse principale à la KU Leuven. En tant que linguiste dans le domaine de l'intelligence artificielle, elle se passionne pour l'amélioration de notre compréhension des modèles de langage grâce à des connaissances linguistiques. Elle explore le traitement automatique du langage naturel dans des contextes multilingues et non anglophones (y compris dans des scénarios à faibles ressources), l'analyse et l'évaluation nuancées de grands modèles linguistiques, ainsi que les technologies de traduction pour les textes et la terminologie spécifiques à un domaine.

Publications

LLMs and Cultural Values: the Impact of Prompt Language and Explicit Cultural Framing
Bram Bulté
Ayla Rigouts Terryn
Large Language Models (LLMs) are rapidly being adopted by users across the globe, who interact with them in a diverse range of languages. At… (voir plus) the same time, there are well-documented imbalances in the training data and optimisation objectives of this technology, raising doubts as to whether LLMs can represent the cultural diversity of their broad user base. In this study, we look at LLMs and cultural values and examine how prompt language and cultural framing influence model responses and their alignment with human values in different countries. We probe 10 LLMs with 63 items from the Hofstede Values Survey Module and World Values Survey, translated into 11 languages, and formulated as prompts with and without different explicit cultural perspectives. Our study confirms that both prompt language and cultural perspective produce variation in LLM outputs, but with an important caveat: While targeted prompting can, to a certain extent, steer LLM responses in the direction of the predominant values of the corresponding countries, it does not overcome the models' systematic bias toward the values associated with a restricted set of countries in our dataset: the Netherlands, Germany, the US, and Japan. All tested models, regardless of their origin, exhibit remarkably similar patterns: They produce fairly neutral responses on most topics, with selective progressive stances on issues such as social tolerance. Alignment with cultural values of human respondents is improved more with an explicit cultural perspective than with a targeted prompt language. Unexpectedly, combining both approaches is no more effective than cultural framing with an English prompt. These findings reveal that LLMs occupy an uncomfortable middle ground: They are responsive enough to changes in prompts to produce variation, but too firmly anchored to specific cultural defaults to adequately represent cultural diversity.
2025-11-05
ArXiv (prépublication)
doi.org
arxiv.org
Introduction to the special issue on Computational Terminology
Ayla Rigouts Terryn
Patrick Drouin
2025-05-23
Terminology. International Journal of Theoretical and Applied Issues in Specialized Communication (publié)
doi.org
Proceedings of the 18th Workshop on Building and Using Comparable Corpora (BUCC)
Serge Sharoff
Ayla Rigouts Terryn
Pierre Zweigenbaum
Reinhard Rapp
2025-01-01
COLING Workshops (publié)
dblp.uni-trier.de
THInC: A Theory-Driven Framework for Computational Humor Detection
Victor De Marez
Thomas Winters
Ayla Rigouts Terryn
Humor is a fundamental aspect of human communication and cognition, as it plays a crucial role in social engagement. Although theories about… (voir plus) humor have evolved over centuries, there is still no agreement on a single, comprehensive humor theory. Likewise, computationally recognizing humor remains a significant challenge despite recent advances in large language models. Moreover, most computational approaches to detecting humor are not based on existing humor theories. This paper contributes to bridging this long-standing gap between humor theory research and computational humor detection by creating an interpretable framework for humor classification, grounded in multiple humor theories, called THInC (Theory-driven Humor Interpretation and Classification). THInC ensembles interpretable GA2M classifiers, each representing a different humor theory. We engineered a transparent flow to actively create proxy features that quantitatively reflect different aspects of theories. An implementation of this framework achieves an F1 score of 0.85. The associative interpretability of the framework enables analysis of proxy efficacy, alignment of joke features with theories, and identification of globally contributing features. This paper marks a pioneering effort in creating a humor detection framework that is informed by diverse humor theories and offers a foundation for future advancements in theory-driven humor classification. It also serves as a first step in automatically comparing humor theories in a quantitative manner.
2024-09-02
ArXiv (prépublication)
doi.org
arxiv.org
THInC: A Theory-Driven Framework for Computational Humor Detection
Victor De Marez
Thomas Winters
Ayla Rigouts Terryn
Humor is a fundamental aspect of human communication and cognition, as it plays a crucial role in social engagement. Although theories about… (voir plus) humor have evolved over centuries, there is still no agreement on a single, comprehensive humor theory. Likewise, computationally recognizing humor remains a significant challenge despite recent advances in large language models. Moreover, most computational approaches to detecting humor are not based on existing humor theories. This paper contributes to bridging this long-standing gap between humor theory research and computational humor detection by creating an interpretable framework for humor classification, grounded in multiple humor theories, called THInC (Theory-driven Humor Interpretation and Classification). THInC ensembles interpretable GA2M classifiers, each representing a different humor theory. We engineered a transparent flow to actively create proxy features that quantitatively reflect different aspects of theories. An implementation of this framework achieves an F1 score of 0.85. The associative interpretability of the framework enables analysis of proxy efficacy, alignment of joke features with theories, and identification of globally contributing features. This paper marks a pioneering effort in creating a humor detection framework that is informed by diverse humor theories and offers a foundation for future advancements in theory-driven humor classification. It also serves as a first step in automatically comparing humor theories in a quantitative manner.
2024-09-02
ArXiv (prépublication)
doi.org
arxiv.org
Exploratory Study on the Impact of English Bias of Generative Large Language Models in Dutch and French
Ayla Rigouts Terryn
Miryam de Lhoneux
The most widely used LLMs like GPT4 and Llama 2 are trained on large amounts of data, mostly in English but are still able to deal with non-… (voir plus)English languages. This English bias leads to lower performance in other languages, especially low-resource ones. This paper studies the linguistic quality of LLMs in two non-English high-resource languages: Dutch and French, with a focus on the influence of English. We first construct a comparable corpus of text generated by humans versus LLMs (GPT-4, Zephyr, and GEITje) in the news domain. We proceed to annotate linguistic issues in the LLM-generated texts, obtaining high inter-annotator agreement, and analyse these annotated issues. We find a substantial influence of English for all models under all conditions: on average, 16% of all annotations of linguistic errors or peculiarities had a clear link to English. Fine-tuning a LLM to a target language (GEITje is fine-tuned on Dutch) reduces the number of linguistic issues and probably also the influence of English. We further find that using a more elaborate prompt leads to linguistically better results than a concise prompt. Finally, increasing the temperature for one of the models leads to lower linguistic quality but does not alter the influence of English.
2024-01-01
HUMEVAL (publié)
www.semanticscholar.org
Exploratory Study on the Impact of English Bias of Generative Large Language Models in Dutch and French
Ayla Rigouts Terryn
Miryam de Lhoneux