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

As part of an automated fact-checking pipeline, the claim veracityclassification task consists in determining if a claim is supported by anassociated piece of evidence. The complexity of gathering labelledclaim-evidence pairs leads to a scarcity of datasets, particularly when dealingwith new domains. In this paper, we introduce SEED, a novel vector-based methodto few-shot claim veracity classification that aggregates pairwise semanticdifferences for claim-evidence pairs. We build on the hypothesis that we cansimulate class representative vectors that capture average semantic differencesfor claim-evidence pairs in a class, which can then be used for classificationof new instances. We compare the performance of our method with competitivebaselines including fine-tuned BERT/RoBERTa models, as well as thestate-of-the-art few-shot veracity classification method that leverageslanguage model perplexity. Experiments conducted on the FEVER and SCIFACTdatasets show consistent improvements over competitive baselines in few-shotsettings. Our code is available.

Aggregating Pairwise Semantic Differences for Few-Shot Claim Veracity Classification