Reproduce “Supporting AI-Augmented Meta-Decision Making with InDecision” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Chance Castañeda, Jessica Mindel, Will Page, Hayden Stec, Manqing Yu, Kenneth Holstein
Researchers from Carnegie Mellon University present InDecision, a mixed-initiative tool and framework for AI-augmented meta-decision making, which supports individuals in iteratively developing and refining their own decision criteria. The system facilitates the surfacing of tacit knowledge and challenges assumptions by having AI act as a "provocateur."
Key insights
- AI can effectively serve as a "provocateur" to stimulate human reflection and surface tacit knowledge during criteria development, rather than merely optimizing outcomes based on given criteria.
- Meta-decision making inherently requires fluid, non-linear iteration between evaluating concrete options and refining abstract criteria, as users spontaneously generate insights across both.
- Users often conceptualize decision criteria with complex interrelationships, such as hierarchies or dependencies, indicating a need for AI tools to support more expressive criteria representations beyond simple lists.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.