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

Fine-tuning a language model on a new domain is standard practice for domainadaptation. However, it can be infeasible when it comes to modern large-scalelanguage models such as GPT-3, which can only be accessed through APIs, makingit difficult to access the internal parameters of the model. In this paper, wepropose $k$NN-Adapter, a method to effectively adapt these black-box largelanguage models (LLMs) to a new domain. The $k$NN-Adapter builds on top of theretrieval-augmented language model, and adaptively learns to interpolate theoutput of the language model with retrieval results from a datastore consistingof the target domain data. Our experiments on four different domainsdemonstrate that $k$NN-Adapter significantly improves perplexity, and worksparticularly well in settings with limited access to LLMs. Additionally, weshow that $k$NN-Adapter is more effective than fine-tuning when the amount oftraining data is limited. We also release a dataset to encourage further study.

$k$NN-Adapter: Efficient Domain Adaptation for Black-Box Language Models