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

Explainable question answering systems should produce not only accurateanswers but also rationales that justify their reasoning and allow humans tocheck their work. But what sorts of rationales are useful and how can we trainsystems to produce them? We propose a new style of rationale for open-bookquestion answering, called \emph{markup-and-mask}, which combines aspects ofextractive and free-text explanations. In the markup phase, the passage isaugmented with free-text markup that enables each sentence to stand on its ownoutside the discourse context. In the masking phase, a sub-span of themarked-up passage is selected. To train a system to produce markup-and-maskrationales without annotations, we leverage in-context learning. Specifically,we generate silver annotated data by sending a series of prompts to a frozenpretrained language model, which acts as a teacher. We then fine-tune a smallerstudent model by training on the subset of rationales that led to correctanswers. The student is "honest" in the sense that it is a pipeline: therationale acts as a bottleneck between the passage and the answer, while the"untrusted" teacher operates under no such constraints. Thus, we offer a newway to build trustworthy pipeline systems from a combination of end-taskannotations and frozen pretrained language models.

Honest Students from Untrusted Teachers: Learning an Interpretable Question-Answering Pipeline from a Pretrained Language Model