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 have demonstrated remarkable few-shot performance, butthe performance can be sensitive to the selection of few-shot instances. Wepropose PATRON, a new method that uses prompt-based uncertainty estimation fordata selection for pre-trained language model fine-tuning under cold-startscenarios, i.e., no initial labeled data are available. In PATRON, we design(1) a prompt-based uncertainty propagation approach to estimate the importanceof data points and (2) a partition-then-rewrite (PTR) strategy to promotesample diversity when querying for annotations. Experiments on six textclassification datasets show that PATRON outperforms the strongest cold-startdata selection baselines by up to 6.9%. Besides, with 128 labels only, PATRONachieves 91.0% and 92.1% of the fully supervised performance based on vanillafine-tuning and prompt-based learning respectively. Our implementation ofPATRON is available at \url{https://github.com/yueyu1030/Patron}.

Cold-Start Data Selection for Few-shot Language Model Fine-tuning: A Prompt-Based Uncertainty Propagation Approach