How Train-Test Leakage Affects Zero-shot Retrieval
Abstract
Neural retrieval models are often trained on (subsets of) the millions of queries of the MS MARCO / ORCAS datasets and then tested on the 250 Robust04 queries or other TREC benchmarks with often only 50 queries. In such setups, many of the few test queries can be very similar to queries from the huge training data -- in fact, 69% of the Robust04 queries have near-duplicates in MS MARCO / ORCAS. We investigate the impact of this unintended train-test leakage by training neural retrieval models on combinations of a fixed number of MS MARCO / ORCAS queries that are highly similar to the actual test queries and an increasing number of other queries. We find that leakage can improve effectiveness and even change the ranking of systems. However, these effects diminish as the amount of leakage among all training instances decreases and thus becomes more realistic.
Cite
@article{arxiv.2206.14759,
title = {How Train-Test Leakage Affects Zero-shot Retrieval},
author = {Maik Fröbe and Christopher Akiki and Martin Potthast and Matthias Hagen},
journal= {arXiv preprint arXiv:2206.14759},
year = {2022}
}
Comments
To appear at the 29th International Symposium on String Processing and Information Retrieval (SPIRE 2022)