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/

Zendy Plus

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

Zendy Tools

Zendy Open

Using the attention map based probing frame-work from (Clark et al., 2019),we observe that, on the RAMS dataset (Ebner et al., 2020), BERT's attentionheads have modest but well above-chance ability to spot event arguments sansany training or domain finetuning, vary-ing from a low of 17.77% for Place to ahigh of 51.61% for Artifact. Next, we find that linear combinations of theseheads, estimated with approx 11% of available total event argument detectionsupervision, can push performance well-higher for some roles - highest twobeing Victim (68.29% Accuracy) and Artifact(58.82% Accuracy). Furthermore, weinvestigate how well our methods do for cross-sentence event arguments. Wepropose a procedure to isolate "best heads" for cross-sentence argumentdetection separately of those for intra-sentence arguments. The heads thusestimated have superior cross-sentence performance compared to their jointlyestimated equivalents, albeit only under the unrealistic assumption that wealready know the argument is present in an-other sentence. Lastly, we seek toisolate to what extent our numbers stem from lexical frequency basedassociations between gold arguments and roles. We propose NONCE, a scheme tocreate adversarial test examples by replacing gold arguments with randomlygenerated "nonce" words. We find that learnt linear combinations are robust toNONCE, though individual best heads can be more sensitive.

BERTering RAMS: What and How Much does BERT Already Know About Event Arguments? -- A Study on the RAMS Dataset