Event
At the Helmholtz AI Conference 2026 (HAICON26), Perelyn's Anton Steuer presented a poster on the question of which cultural values are embedded in Large Language Models. Unlike the previously common approach of presenting models with questionnaires from cultural studies and evaluating their responses, his approach measures word associations directly in the embedding space, i.e., in the model's internal representations.
Across six LLMs from four countries, a clear tendency towards Western-influenced values is evident.
Perelyn is eager to drive and support such excellent research initiatives, since companies increasingly rely on LLMs and can only scale them responsibly once they understand which cultural values and biases are embedded in these models.
Here the link to the poster
Event
Liliya Imasheva presented a validation pipeline for evaluating AI summaries at Conf42 Large Language Models 2026.
Event
At the IOCMA 2026 conference, Perelyn presented our method, which transforms graph data such as supply chains or networks so that AI models can learn more reliably from it, even over long distances.
News
The contributions from the "Knowledge Management" workshop at KI2025 have been published as a collected volume. Among them is a paper by our team on a novel knowledge graph-based RAG method, developed to provide accurate, verifiable answers with fewer errors and lower costs.
