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

Recent work has shown the promise of learning with human feedback paradigmsto produce human-determined high-quality text. Existing works use humanfeedback to train large language models (LLMs) in general domain abstractivesummarization and have obtained summary quality exceeding traditionallikelihood training. In this paper, we focus on a less explored form of humanfeedback -- Human Edits. We propose Sequence Alignment (un)Likelihood Training(SALT), a novel technique to use both the human-edited and model-generated datatogether in the training loop. In addition, we demonstrate simulating HumanEdits with ground truth summaries coming from existing training data --Imitation edits, along with the model-generated summaries obtained after thetraining, to reduce the need for expensive human-edit data. In our experiments,we extend human feedback exploration from general domain summarization tomedical domain summarization. Our results demonstrate the effectiveness of SALTin improving the summary quality with Human and Imitation Edits. Throughadditional experiments, we show that SALT outperforms the conventional RLHFmethod (designed for human preferences) -- DPO, when applied to human-editdata. We hope the evidence in our paper prompts researchers to explore,collect, and better use different human feedback approaches scalably.

Improving Summarization with Human Edits