English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Recent progress in information retrieval finds that embedding query anddocument representation into multi-vector yields a robust bi-encoder retrieveron out-of-distribution datasets. In this paper, we explore whether lateinteraction, the simplest form of multi-vector, is also helpful to neuralrerankers that only use the [CLS] vector to compute the similarity score.Although intuitively, the attention mechanism of rerankers at the previouslayers already gathers the token-level information, we find adding lateinteraction still brings an extra 5% improvement in average onout-of-distribution datasets, with little increase in latency and nodegradation in in-domain effectiveness. Through extensive experiments andanalysis, we show that the finding is consistent across different model sizesand first-stage retrievers of diverse natures and that the improvement is moreprominent on longer queries.

Improving Out-of-Distribution Generalization of Neural Rerankers with Contextualized Late Interaction