Large Language Models (LLMs) have revolutionized natural language understanding and generation, driving scientific research forward by assisting in all steps of the scientific process, ranging from identifying research gaps to accelerating complex data analysis. In our interdisciplinary research, we explore the diverse applications of LLMs across various scientific domains, from meta-analyses in the science of science to practical implementations in applied fields.
Papers
PRESENTATIONS
Memorization vs. Reasoning: What the vec happens inside LLMs (when solving math questions)?
Presented at the Junior Colloquium UCL on November 27, 2024.
Large Language Models Reflect Human Citation Patterns with a Heightened Citation Bias
Andres Algaba, Carmen Mazijn, Vincent Holst, Floriano Tori, Sylvia Wenmackers & Vincent Ginis
Presented at the NAACL AI & Scientific Discovery Workshop on May 3, 2025 in Albuquerque.
Posters
Large Language Models Reflect Human Citation Patterns with a Heightened Citation Bias
Andres Algaba, Carmen Mazijn, Vincent Holst, Floriano Tori, Sylvia Wenmackers & Vincent Ginis
Presented at NAACL on May 1, 2025 in Albuquerque.
How Deeply Do LLMs Internalize Human Citation Practices? A Graph-Structural and Embedding-Based Evaluation
Melika Mobini, Vincent Holst, Floriano Tori, Andres Algaba & Vincent Ginis
Presented at the ICLR 2025 Workshop on Human-AI Coevolution (HAIC) on April 26, 2025 in Singapore.