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

Moral or ethical judgments rely heavily on the specific contexts in whichthey occur. Understanding varying shades of defeasible contextualizations(i.e., additional information that strengthens or attenuates the moralacceptability of an action) is critical to accurately represent the subtletyand intricacy of grounded human moral judgment in real-life scenarios.  We introduce defeasible moral reasoning: a task to provide grounded contextsthat make an action more or less morally acceptable, along with commonsenserationales that justify the reasoning. To elicit high-quality task data, wetake an iterative self-distillation approach that starts from a small amount ofunstructured seed knowledge from GPT-3 and then alternates between (1)self-distillation from student models; (2) targeted filtering with a criticmodel trained by human judgment (to boost validity) and NLI (to boostdiversity); (3) self-imitation learning (to amplify the desired data quality).This process yields a student model that produces defeasible contexts withimproved validity, diversity, and defeasibility. From this model we distill ahigh-quality dataset, \delta-Rules-of-Thumb, of 1.2M entries ofcontextualizations and rationales for 115K defeasible moral actions ratedhighly by human annotators 85.9% to 99.8% of the time. Using \delta-RoT weobtain a final student model that wins over all intermediate student models bya notable margin.

What Makes it Ok to Set a Fire? Iterative Self-distillation of Contexts and Rationales for Disambiguating Defeasible Social and Moral Situations