Elena Kramer
My research investigates the meta-languages used to express knowledge in Mathematics and Computer Science, focusing on the linguistic structures that underpin definitions, axioms, proofs, algorithms, and heuristics. By analyzing large corpora of academic texts from multiple subfields using clustering techniques and Transformer-based language models such as XLNet, the study identifies shared meta-linguistic patterns across the two disciplines. The findings aim to support knowledge transfer by enabling educational content to be automatically rephrased into more familiar meta-languages, thereby improving comprehension and accessibility for learners.
Ongoing research extends this meta-linguistic framework to the emerging field of multi-agent AI systems by investigating the languages used for communication between AI agents. The goal is to develop methods for analyzing, evaluating, and optimizing inter-agent communication to improve coordination, knowledge exchange, and collaborative decision-making in complex multi-agent environments.
