We are pleased to announce that our latest research article has been published in the journal Discover Computing. The publication marks another important milestone in our joint research with Nokia Bell Labs.

The co-authors of the publication are: Csaba Kiss (PhD student) and Dr. Marcell Nagy (HSDSLab), as well as Péter Szilágyi (Nokia Bell Labs). The study introduces a new method called “Max–Min Semantic Chunking,” which combines semantic similarity and a Max–Min algorithm to identify highly cohesive semantic units in texts.

This approach proves particularly effective in Retrieval Augmented Generation (RAG) systems, improving their accuracy and contextual sensitivity. The method opens new possibilities for the development of AI-based text understanding and knowledge management.

Congratulations to the authors on the publication, and thank you to our partners at Nokia Bell Labs for the inspiring collaboration!

The full article is available here: https://link.springer.com/article/10.1007/s10791-025-09638-7