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

Answering compositional questions that require multiple steps of reasoningagainst text is challenging, especially when they involve discrete, symbolicoperations. Neural module networks (NMNs) learn to parse such questions asexecutable programs composed of learnable modules, performing well on syntheticvisual QA domains. However, we find that it is challenging to learn thesemodels for non-synthetic questions on open-domain text, where a model needs todeal with the diversity of natural language and perform a broader range ofreasoning. We extend NMNs by: (a) introducing modules that reason over aparagraph of text, performing symbolic reasoning (such as arithmetic, sorting,counting) over numbers and dates in a probabilistic and differentiable manner;and (b) proposing an unsupervised auxiliary loss to help extract argumentsassociated with the events in text. Additionally, we show that a limited amountof heuristically-obtained question program and intermediate module outputsupervision provides sufficient inductive bias for accurate learning. Ourproposed model significantly outperforms state-of-the-art models on a subset ofthe DROP dataset that poses a variety of reasoning challenges that are coveredby our modules.

Neural Module Networks for Reasoning over Text