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

Despite the success of language models using neural networks, it remainsunclear to what extent neural models have the generalization ability to performinferences. In this paper, we introduce a method for evaluating whether neuralmodels can learn systematicity of monotonicity inference in natural language,namely, the regularity for performing arbitrary inferences with generalizationon composition. We consider four aspects of monotonicity inferences and testwhether the models can systematically interpret lexical and logical phenomenaon different training/test splits. A series of experiments show that threeneural models systematically draw inferences on unseen combinations of lexicaland logical phenomena when the syntactic structures of the sentences aresimilar between the training and test sets. However, the performance of themodels significantly decreases when the structures are slightly changed in thetest set while retaining all vocabularies and constituents already appearing inthe training set. This indicates that the generalization ability of neuralmodels is limited to cases where the syntactic structures are nearly the sameas those in the training set.

Do Neural Models Learn Systematicity of Monotonicity Inference in Natural Language?