How Different are Pre-trained Transformers for Text Ranking?
This work helps understand neural rankers relative to traditional ones, addressing a specific problem in information retrieval research, but it is incremental as it builds on existing models and datasets.
The paper analyzed BERT-based cross-encoders versus BM25 for passage retrieval, finding substantial differences in relevance notions that highlight strengths and weaknesses of BERT, with results based on the MS Marco/TREC Deep Learning Track setup.
In recent years, large pre-trained transformers have led to substantial gains in performance over traditional retrieval models and feedback approaches. However, these results are primarily based on the MS Marco/TREC Deep Learning Track setup, with its very particular setup, and our understanding of why and how these models work better is fragmented at best. We analyze effective BERT-based cross-encoders versus traditional BM25 ranking for the passage retrieval task where the largest gains have been observed, and investigate two main questions. On the one hand, what is similar? To what extent does the neural ranker already encompass the capacity of traditional rankers? Is the gain in performance due to a better ranking of the same documents (prioritizing precision)? On the other hand, what is different? Can it retrieve effectively documents missed by traditional systems (prioritizing recall)? We discover substantial differences in the notion of relevance identifying strengths and weaknesses of BERT that may inspire research for future improvement. Our results contribute to our understanding of (black-box) neural rankers relative to (well-understood) traditional rankers, help understand the particular experimental setting of MS-Marco-based test collections.