iiinova — where your digital self keeps evolving

Leijie Wang

@leijiew.bsky.social

A third-year PhD student at the University of Washington

Leijie Wang @leijiew.bsky.social · 25/03/2025

Huge thanks to my wonderful collaborators Kathryn Yurechko, Pranati Dani, @cqz.bsky.social and @axz.bsky.social. Full details here ➡️ arxiv.org/pdf/2409.03247

arxiv.org

020

Leijie Wang @leijiew.bsky.social · 25/03/2025

All three strategies struggled with iterative refinement. Interestingly, participants adopted hybrid approaches when iterating on their prompt filters – like providing examples as in-context examples or writing rule-like prompts.

100

Leijie Wang @leijiew.bsky.social · 25/03/2025

Despite 🤖LLM prompting’s better performance, participants preferred mixed strategies to create their filters. For example, when their preferences were ill-defined but intuitive, 🔎labeling examples was considered the easiest way. (🧵4/N)

100

Leijie Wang @leijiew.bsky.social · 25/03/2025

To answer this question, our study had 37 non-programmers create personal content filters using these three strategies. (🧵3/N)

100

Leijie Wang @leijiew.bsky.social · 25/03/2025

Existing content filter tools often expect lay people to work in a single, long setup session. Yet users engage with social media in short, everyday sessions. How can we support social media users to more easily create and iterate on their filters? (🧵2/N)

100

Leijie Wang @leijiew.bsky.social · 25/03/2025

Can LLM prompting help social media users create and iterate on their content filters more easily? In our #CHI2025 paper, we compared in an experiment three authoring strategies: 🤖 Prompting LLM 🔎 Labeling examples for ML classifiers 📐 Authoring keyword rules (🧵1/N)

1226