Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Directed Diversity: Leveraging Language Embedding Distances for Collective Creativity in Crowd Ideation

About

Crowdsourcing can collect many diverse ideas by prompting ideators individually, but this can generate redundant ideas. Prior methods reduce redundancy by presenting peers' ideas or peer-proposed prompts, but these require much human coordination. We introduce Directed Diversity, an automatic prompt selection approach that leverages language model embedding distances to maximize diversity. Ideators can be directed towards diverse prompts and away from prior ideas, thus improving their collective creativity. Since there are diverse metrics of diversity, we present a Diversity Prompting Evaluation Framework consolidating metrics from several research disciplines to analyze along the ideation chain - prompt selection, prompt creativity, prompt-ideation mediation, and ideation creativity. Using this framework, we evaluated Directed Diversity in a series of a simulation study and four user studies for the use case of crowdsourcing motivational messages to encourage physical activity. We show that automated diverse prompting can variously improve collective creativity across many nuanced metrics of diversity.

Samuel Rhys Cox, Yunlong Wang, Ashraf Abdul, Christian von der Weth, Brian Y. Lim• 2021

Related benchmarks

TaskDatasetResultRank
Sentiment ClassificationTwitter Financial News (test)
F1 Score0.542
23
Text2SQLBIRD Movies
Execution Accuracy46.9
9
Image Classificationunmet-promise (Split 1)
Task Performance57
9
Text2SQLBIRD Computer Students
Execution Accuracy48.2
9
Text2SQLBIRD App Store
Execution Accuracy33.9
9
Image Classificationunmet-promise (Split 2)
Accuracy54.8
9
Web navigationWebNav
Task Performance25.8
9
Image Classificationunmet-promise (Split 3)
Task Performance52.2
9
Sentiment AnalysisSentiment Dataset
Spearman Correlation0.68
8
Text-to-SQLBIRD Movies
Spearman Correlation0.61
8
Showing 10 of 16 rows

Other info

Follow for update