The Expando-Mono-Duo Design Pattern for Text Ranking with Pretrained Sequence-to-Sequence Models
This work addresses text retrieval challenges for information retrieval practitioners by offering a validated, open-sourced design pattern, though it is incremental as it builds on existing multi-stage ranking architectures.
The paper tackles text ranking problems by proposing the Expando-Mono-Duo design pattern, which uses pretrained sequence-to-sequence models in a multi-stage architecture with document expansion and reranking components, achieving state-of-the-art or near state-of-the-art effectiveness on tasks like MS MARCO and TREC-COVID, including in some zero-shot settings.
We propose a design pattern for tackling text ranking problems, dubbed "Expando-Mono-Duo", that has been empirically validated for a number of ad hoc retrieval tasks in different domains. At the core, our design relies on pretrained sequence-to-sequence models within a standard multi-stage ranking architecture. "Expando" refers to the use of document expansion techniques to enrich keyword representations of texts prior to inverted indexing. "Mono" and "Duo" refer to components in a reranking pipeline based on a pointwise model and a pairwise model that rerank initial candidates retrieved using keyword search. We present experimental results from the MS MARCO passage and document ranking tasks, the TREC 2020 Deep Learning Track, and the TREC-COVID challenge that validate our design. In all these tasks, we achieve effectiveness that is at or near the state of the art, in some cases using a zero-shot approach that does not exploit any training data from the target task. To support replicability, implementations of our design pattern are open-sourced in the Pyserini IR toolkit and PyGaggle neural reranking library.