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

Neural rankers based on deep pretrained language models (LMs) have been shownto improve many information retrieval benchmarks. However, these methods areaffected by their the correlation between pretraining domain and target domainand rely on massive fine-tuning relevance labels. Directly applying pretrainingmethods to specific domains may result in suboptimal search quality becausespecific domains may have domain adaption problems, such as the COVID domain.This paper presents a search system to alleviate the special domain adaptionproblem. The system utilizes the domain-adaptive pretraining and few-shotlearning technologies to help neural rankers mitigate the domain discrepancyand label scarcity problems. Besides, we also integrate dense retrieval toalleviate traditional sparse retrieval's vocabulary mismatch obstacle. Oursystem performs the best among the non-manual runs in Round 2 of the TREC-COVIDtask, which aims to retrieve useful information from scientific literaturerelated to COVID-19. Our code is publicly available athttps://github.com/thunlp/OpenMatch.

CMT in TREC-COVID Round 2: Mitigating the Generalization Gaps from Web to Special Domain Search