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

Knowledge-intensive language tasks (KILTs) benefit from retrievinghigh-quality relevant contexts from large external knowledge corpora. Learningtask-specific retrievers that return relevant contexts at an appropriate levelof semantic granularity, such as a document retriever, passage retriever,sentence retriever, and entity retriever, may help to achieve betterperformance on the end-to-end task. But a task-specific retriever usually haspoor generalization ability to new domains and tasks, and it may be costly todeploy a variety of specialised retrievers in practice. We propose a unifiedgenerative retriever (UGR) that combines task-specific effectiveness withrobust performance over different retrieval tasks in KILTs. To achieve thisgoal, we make two major contributions: (i) To unify different retrieval tasksinto a single generative form, we introduce an n-gram-based identifier forrelevant contexts at different levels of granularity in KILTs. And (ii) toaddress different retrieval tasks with a single model, we employ a promptlearning strategy and investigate three methods to design prompt tokens foreach task. In this way, the proposed UGR model can not only share commonknowledge across tasks for better generalization, but also perform differentretrieval tasks effectively by distinguishing task-specific characteristics. Wetrain UGR on a heterogeneous set of retrieval corpora with well-designedprompts in a supervised and multi-task fashion. Experimental results on theKILT benchmark demonstrate the effectiveness of UGR on in-domain datasets,out-of-domain datasets, and unseen tasks.

A Unified Generative Retriever for Knowledge-Intensive Language Tasks via Prompt Learning