Preprint
Aug 2026
Robustness of IR Models to Collection Growth
This study empirically evaluates the robustness of an IR model to the addition of non-relevant documents by merging two collections with negligible topic overlap and finds that MDA is more effective than MDD for retrieval, whereas MDD and MDA rerankers are equally effective.
Emmanouil Georgios Lionis, Sean MacAvaney, Debasis Ganguly
· 0 citations