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

Deep pre-trained contextualized encoders like BERT (Delvin et al., 2019)demonstrate remarkable performance on a range of downstream tasks. A recentline of research in probing investigates the linguistic knowledge implicitlylearned by these models during pre-training. While most work in probingoperates on the task level, linguistic tasks are rarely uniform and can berepresented in a variety of formalisms. Any linguistics-based probing studythereby inevitably commits to the formalism used to annotate the underlyingdata. Can the choice of formalism affect probing results? To investigate, weconduct an in-depth cross-formalism layer probing study in role semantics. Wefind linguistically meaningful differences in the encoding of semantic role-and proto-role information by BERT depending on the formalism and demonstratethat layer probing can detect subtle differences between the implementations ofthe same linguistic formalism. Our results suggest that linguistic formalism isan important dimension in probing studies, along with the commonly usedcross-task and cross-lingual experimental settings.

A Matter of Framing: The Impact of Linguistic Formalism on Probing Results