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

This paper describes Difference-aware Deep continuous prompt for ContrastiveSentence Embeddings (D2CSE) that learns sentence embeddings. Compared tostate-of-the-art approaches, D2CSE computes sentence vectors that areexceptional to distinguish a subtle difference in similar sentences byemploying a simple neural architecture for continuous prompts. Unlike existingarchitectures that require multiple pretrained language models (PLMs) toprocess a pair of the original and corrupted (subtly modified) sentences, D2CSEavoids cumbersome fine-tuning of multiple PLMs by only optimizing continuousprompts by performing multiple tasks -- i.e., contrastive learning andconditional replaced token detection all done in a self-guided manner. D2CSEoverloads a single PLM on continuous prompts and greatly saves memoryconsumption as a result. The number of training parameters in D2CSE is reducedto about 1\% of existing approaches while substantially improving the qualityof sentence embeddings. We evaluate D2CSE on seven Semantic Textual Similarity(STS) benchmarks, using three different metrics, namely, Spearman's rankcorrelation, recall@K for a retrieval task, and the anisotropy of an embeddingspace measured in alignment and uniformity. Our empirical results suggest thatshallow (not too meticulously devised) continuous prompts can be honedeffectively for multiple NLP tasks and lead to improvements upon existingstate-of-the-art approaches.

D2CSE: Difference-aware Deep continuous prompts for Contrastive Sentence Embeddings