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

Weakly supervised question answering usually has only the final answers assupervision signals while the correct solutions to derive the answers are notprovided. This setting gives rise to the spurious solution problem: there mayexist many spurious solutions that coincidentally derive the correct answer,but training on such solutions can hurt model performance (e.g., producingwrong solutions or answers). For example, for discrete reasoning tasks as onDROP, there may exist many equations to derive a numeric answer, and typicallyonly one of them is correct. Previous learning methods mostly filter outspurious solutions with heuristics or using model confidence, but do notexplicitly exploit the semantic correlations between a question and itssolution. In this paper, to alleviate the spurious solution problem, we proposeto explicitly exploit such semantic correlations by maximizing the mutualinformation between question-answer pairs and predicted solutions. Extensiveexperiments on four question answering datasets show that our methodsignificantly outperforms previous learning methods in terms of taskperformance and is more effective in training models to produce correctsolutions.

A Mutual Information Maximization Approach for the Spurious Solution Problem in Weakly Supervised Question Answering