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

Overlapping frequently occurs in paired texts in natural language processingtasks like text editing and semantic similarity evaluation. Better evaluationof the semantic distance between the overlapped sentences benefits the languagesystem's understanding and guides the generation. Since conventional semanticmetrics are based on word representations, they are vulnerable to thedisturbance of overlapped components with similar representations. This paperaims to address the issue with a mask-and-predict strategy. We take the wordsin the longest common sequence (LCS) as neighboring words and use maskedlanguage modeling (MLM) from pre-trained language models (PLMs) to predict thedistributions on their positions. Our metric, Neighboring DistributionDivergence (NDD), represent the semantic distance by calculating the divergencebetween distributions in the overlapped parts. Experiments on Semantic TextualSimilarity show NDD to be more sensitive to various semantic differences,especially on highly overlapped paired texts. Based on the discovery, wefurther implement an unsupervised and training-free method for textcompression, leading to a significant improvement on the previousperplexity-based method. The high scalability of our method even enables NDD tooutperform the supervised state-of-the-art in domain adaption by a huge margin.Further experiments on syntax and semantics analyses verify the awareness ofinternal sentence structures, indicating the high potential of NDD for furtherstudies.

Contextualized Semantic Distance between Highly Overlapped Texts