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

Ultra-fine entity typing (UFET) aims to predict a wide range of type phrasesthat correctly describe the categories of a given entity mention in a sentence.Most recent works infer each entity type independently, ignoring thecorrelations between types, e.g., when an entity is inferred as a president, itshould also be a politician and a leader. To this end, we use an undirectedgraphical model called pairwise conditional random field (PCRF) to formulatethe UFET problem, in which the type variables are not only unarily influencedby the input but also pairwisely relate to all the other type variables. We usevarious modern backbones for entity typing to compute unary potentials, andderive pairwise potentials from type phrase representations that both captureprior semantic information and facilitate accelerated inference. We usemean-field variational inference for efficient type inference on very largetype sets and unfold it as a neural network module to enable end-to-endtraining. Experiments on UFET show that the Neural-PCRF consistentlyoutperforms its backbones with little cost and results in a competitiveperformance against cross-encoder based SOTA while being thousands of timesfaster. We also find Neural- PCRF effective on a widely used fine-grainedentity typing dataset with a smaller type set. We pack Neural-PCRF as a networkmodule that can be plugged onto multi-label type classifiers with ease andrelease it in https://github.com/modelscope/adaseq/tree/master/examples/NPCRF.

Modeling Label Correlations for Ultra-Fine Entity Typing with Neural Pairwise Conditional Random Field