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

Semantic representation learning for sentences is an important andwell-studied problem in NLP. The current trend for this task involves traininga Transformer-based sentence encoder through a contrastive objective with text,i.e., clustering sentences with semantically similar meanings and scatteringothers. In this work, we find the performance of Transformer models as sentenceencoders can be improved by training with multi-modal multi-task losses, usingunpaired examples from another modality (e.g., sentences and unrelatedimage/audio data). In particular, besides learning by the contrastive loss ontext, our model clusters examples from a non-linguistic domain (e.g.,visual/audio) with a similar contrastive loss at the same time. The reliance ofour framework on unpaired non-linguistic data makes it language-agnostic,enabling it to be widely applicable beyond English NLP. Experiments on 7semantic textual similarity benchmarks reveal that models trained with theadditional non-linguistic (images/audio) contrastive objective lead to higherquality sentence embeddings. This indicates that Transformer models are able togeneralize better by doing a similar task (i.e., clustering) with unpairedexamples from different modalities in a multi-task fashion.

Non-Linguistic Supervision for Contrastive Learning of Sentence Embeddings