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

Transformers have been shown to be able to perform deductive reasoning on alogical rulebase containing rules and statements written in English naturallanguage. While the progress is promising, it is currently unclear if thesemodels indeed perform logical reasoning by understanding the underlying logicalsemantics in the language. To this end, we propose RobustLR, a suite ofevaluation datasets that evaluate the robustness of these models to minimallogical edits in rulebases and some standard logical equivalence conditions. Inour experiments with RoBERTa and T5, we find that the models trained in priorworks do not perform consistently on the different perturbations in RobustLR,thus showing that the models are not robust to the proposed logicalperturbations. Further, we find that the models find it especially hard tolearn logical negation and disjunction operators. Overall, using our evaluationsets, we demonstrate some shortcomings of the deductive reasoning-basedlanguage models, which can eventually help towards designing better models forlogical reasoning over natural language. All the datasets and code base havebeen made publicly available.

RobustLR: Evaluating Robustness to Logical Perturbation in Deductive Reasoning