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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Machine Reading Comprehension (MRC) is the task of answering a question overa paragraph of text. While neural MRC systems gain popularity and achievenoticeable performance, issues are being raised with the methodology used toestablish their performance, particularly concerning the data design of goldstandards that are used to evaluate them. There is but a limited understandingof the challenges present in this data, which makes it hard to draw comparisonsand formulate reliable hypotheses. As a first step towards alleviating theproblem, this paper proposes a unifying framework to systematically investigatethe present linguistic features, required reasoning and background knowledgeand factual correctness on one hand, and the presence of lexical cues as alower bound for the requirement of understanding on the other hand. We proposea qualitative annotation schema for the first and a set of approximativemetrics for the latter. In a first application of the framework, we analysemodern MRC gold standards and present our findings: the absence of featuresthat contribute towards lexical ambiguity, the varying factual correctness ofthe expected answers and the presence of lexical cues, all of which potentiallylower the reading comprehension complexity and quality of the evaluation data.

A Framework for Evaluation of Machine Reading Comprehension Gold Standards