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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

We probe pre-trained transformer language models for bridging inference. Wefirst investigate individual attention heads in BERT and observe that attentionheads at higher layers prominently focus on bridging relations in-comparisonwith the lower and middle layers, also, few specific attention headsconcentrate consistently on bridging. More importantly, we consider languagemodels as a whole in our second approach where bridging anaphora resolution isformulated as a masked token prediction task (Of-Cloze test). Our formulationproduces optimistic results without any fine-tuning, which indicates thatpre-trained language models substantially capture bridging inference. Ourfurther investigation shows that the distance between anaphor-antecedent andthe context provided to language models play an important role in theinference.

Probing for Bridging Inference in Transformer Language Models