Litcius/Paper detail

TILDE: Term Independent Likelihood moDEl for Passage Re-ranking

Shengyao Zhuang, Guido Zuccon

202184 citationsDOI

Abstract

Deep language models (deep LMs) are increasingly being used for full text retrieval or within cascade retrieval pipelines as later-stage re-rankers. A problem with using deep LMs is that, at query time, a slow inference step needs to be performed -- this hinders the practical adoption of these powerful retrieval models, or limits sensibly how many documents can be considered for re-ranking.

Topics & Concepts

Ranking (information retrieval)Computer scienceTerm (time)InferenceArtificial intelligenceCascadeInformation retrievalMachine learningChromatographyPhysicsQuantum mechanicsChemistryTopic ModelingMachine Learning and AlgorithmsData Quality and Management
TILDE: Term Independent Likelihood moDEl for Passage Re-ranking | Litcius