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

Dual-encoder-based neural retrieval models achieve appreciable performanceand complement traditional lexical retrievers well due to their semanticmatching capabilities, which makes them a common choice for hybrid IR systems.However, these models exhibit a performance bottleneck in the online queryencoding step, as the corresponding query encoders are usually large andcomplex Transformer models.  In this paper we investigate heterogeneous dual-encoder models, where the twoencoders are separate models that do not share parameters or initializations.We empirically show that heterogeneous dual-encoders are susceptible tocollapsing representations, causing them to output constant trivialrepresentations when they are fine-tuned using a standard contrastive loss dueto a distribution mismatch. We propose DAFT, a simple two-stage fine-tuningapproach that aligns the two encoders in order to prevent them from collapsing.We further demonstrate how DAFT can be used to train efficient heterogeneousdual-encoder models using lightweight query encoders.

Distribution-Aligned Fine-Tuning for Efficient Neural Retrieval