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

Open Domain Question Answering (ODQA) on a large-scale corpus of documents(e.g. Wikipedia) is a key challenge in computer science. Althoughtransformer-based language models such as Bert have shown on SQuAD the abilityto surpass humans for extracting answers in small passages of text, they sufferfrom their high complexity when faced to a much larger search space. The mostcommon way to tackle this problem is to add a preliminary Information Retrievalstep to heavily filter the corpus and only keep the relevant passages. In thispaper, we propose a more direct and complementary solution which consists inapplying a generic change in the architecture of transformer-based models todelay the attention between subparts of the input and allow a more efficientmanagement of computations. The resulting variants are competitive with theoriginal models on the extractive task and allow, on the ODQA setting, asignificant speedup and even a performance improvement in many cases.

Delaying Interaction Layers in Transformer-based Encoders for Efficient Open Domain Question Answering