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

State-of-the-art neural models typically encode document-query pairs usingcross-attention for re-ranking. To this end, models generally utilize anencoder-only (like BERT) paradigm or an encoder-decoder (like T5) approach.These paradigms, however, are not without flaws, i.e., running the model on allquery-document pairs at inference-time incurs a significant computational cost.This paper proposes a new training and inference paradigm for re-ranking. Wepropose to finetune a pretrained encoder-decoder model using in the form ofdocument to query generation. Subsequently, we show that this encoder-decoderarchitecture can be decomposed into a decoder-only language model duringinference. This results in significant inference time speedups since thedecoder-only architecture only needs to learn to interpret static encoderembeddings during inference. Our experiments show that this new paradigmachieves results that are comparable to the more expensive cross-attentionranking approaches while being up to 6.8X faster. We believe this work pavesthe way for more efficient neural rankers that leverage large pretrainedmodels.

ED2LM: Encoder-Decoder to Language Model for Faster Document Re-ranking Inference