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

Tailoring outputs of large language models, such as ChatGPT, to specific userneeds remains a challenge despite their impressive generation quality. In thispaper, we propose a tri-agent generation pipeline consisting of a generator, aninstructor, and an editor to enhance the customization of generated outputs.The generator produces an initial output, the user-specific instructorgenerates editing instructions, and the editor generates a revised outputaligned with user preferences. The inference-only large language model(ChatGPT) serves as both the generator and the editor, while a smaller modelacts as the user-specific instructor to guide the generation process towarduser needs. The instructor is trained using editor-steered reinforcementlearning, leveraging feedback from the large-scale editor model to optimizeinstruction generation. Experimental results on two abstractive summarizationdatasets demonstrate the effectiveness of our approach in generating outputsthat better fulfill user expectations.

ChatGPT-steered Editing Instructor for Customization of Abstractive Summarization