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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Entity resolution (ER) is an important data integration task with a widespectrum of applications. The state-of-the-art solutions on ER rely onpre-trained language models (PLMs), which require fine-tuning on a lot oflabeled matching/non-matching entity pairs. Recently, large languages models(LLMs), such as GPT-4, have shown the ability to perform many tasks withouttuning model parameters, which is known as in-context learning (ICL) thatfacilitates effective learning from a few labeled input context demonstrations.However, existing ICL approaches to ER typically necessitate providing a taskdescription and a set of demonstrations for each entity pair and thus havelimitations on the monetary cost of interfacing LLMs. To address the problem,in this paper, we provide a comprehensive study to investigate how to develop acost-effective batch prompting approach to ER. We introduce a framework BATCHERconsisting of demonstration selection and question batching and exploredifferent design choices that support batch prompting for ER. We also devise acovering-based demonstration selection strategy that achieves an effectivebalance between matching accuracy and monetary cost. We conduct a thoroughevaluation to explore the design space and evaluate our proposed strategies.Through extensive experiments, we find that batch prompting is verycost-effective for ER, compared with not only PLM-based methods fine-tuned withextensive labeled data but also LLM-based methods with manually designedprompting. We also provide guidance for selecting appropriate design choicesfor batch prompting.

Cost-Effective In-Context Learning for Entity Resolution: A Design Space Exploration