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

Natural Language Sentence Matching (NLSM) has gained substantial attentionfrom both academics and the industry, and rich public datasets contribute a lotto this process. However, biased datasets can also hurt the generalizationperformance of trained models and give untrustworthy evaluation results. Formany NLSM datasets, the providers select some pairs of sentences into thedatasets, and this sampling procedure can easily bring unintended pattern,i.e., selection bias. One example is the QuoraQP dataset, where somecontent-independent naive features are unreasonably predictive. Such featuresare the reflection of the selection bias and termed as the leakage features. Inthis paper, we investigate the problem of selection bias on six NLSM datasetsand find that four out of them are significantly biased. We further propose atraining and evaluation framework to alleviate the bias. Experimental resultson QuoraQP suggest that the proposed framework can improve the generalizationability of trained models, and give more trustworthy evaluation results forreal-world adoptions.

Selection Bias Explorations and Debias Methods for Natural Language Sentence Matching Datasets