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

Large language models (LLMs) have demonstrated impressive capabilities invarious natural language processing tasks. Despite this, their application toinformation retrieval (IR) tasks is still challenging due to the infrequentoccurrence of many IR-specific concepts in natural language. While prompt-basedmethods can provide task descriptions to LLMs, they often fall short infacilitating comprehensive understanding and execution of IR tasks, therebylimiting LLMs' applicability. To address this gap, in this work, we explore thepotential of instruction tuning to enhance LLMs' proficiency in IR tasks. Weintroduce a novel instruction tuning dataset, INTERS, encompassing 21 tasksacross three fundamental IR categories: query understanding, documentunderstanding, and query-document relationship understanding. The data arederived from 43 distinct datasets with manually written templates. Ourempirical results reveal that INTERS significantly boosts the performance ofvarious publicly available LLMs, such as LLaMA, Mistral, and Phi, insearch-related tasks. Furthermore, we conduct a comprehensive analysis toascertain the effects of base model selection, instruction design, volume ofinstructions, and task variety on performance. We make our dataset and themodels fine-tuned on it publicly accessible at https://github.com/DaoD/INTERS.

INTERS: Unlocking the Power of Large Language Models in Search with Instruction Tuning