معرفی
Dr. Nicola Tonellotto serves as Honorary Research Fellow at the University of Glasgow's School of Computing Science. His research advances information retrieval through neural approaches and efficiency optimization. Recent work includes embedding pruning for large-scale recommendation systems and reproducibility studies of retrieval frameworks.
His investigations focus on constant-space multi-vector retrieval techniques, dynamic optimization of sub-item embeddings, and cross-encoder refinement through multi-stage fine-tuning. Collaborations explore document quality scoring for web crawling and model updating strategies for document streams.
Dr. Tonellotto contributes to PyTerrier framework development and maintains active research partnerships in retrieval system optimization. His group examines practical implementations of dense retrieval models under resource constraints.
