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

Multi-hop question answering (QA) requires a model to retrieve and integrateinformation from different parts of a long text to answer a question. Humansanswer this kind of complex questions via a divide-and-conquer approach. Inthis paper, we investigate whether top-performing models for multi-hopquestions understand the underlying sub-questions like humans. We adopt aneural decomposition model to generate sub-questions for a multi-hop complexquestion, followed by extracting the corresponding sub-answers. We show thatmultiple state-of-the-art multi-hop QA models fail to correctly answer a largeportion of sub-questions, although their corresponding multi-hop questions arecorrectly answered. This indicates that these models manage to answer themulti-hop questions using some partial clues, instead of truly understandingthe reasoning paths. We also propose a new model which significantly improvesthe performance on answering the sub-questions. Our work takes a step forwardtowards building a more explainable multi-hop QA system.

Do Multi-Hop Question Answering Systems Know How to Answer the Single-Hop Sub-Questions?