Internal Pluralism and the Limits of Pairwise Comparisons
Local pairwise comparisons are a standard tool for learning how people want decision rules to work, e.g., in participatory design or alignment. However, their use builds in two strong assumptions: that local comparisons are sufficient evidence about how a person wants an automate...
ArXiv CS.AI
Source
Generated insight by OnAirToday AI
This research highlights a critical limitation of using pairwise comparisons to understand human preferences for AI decision-making. It reveals how internal pluralism, where individuals hold multiple authoritative priorities, can lead to inaccurate or incomplete preference learning if not accounted for.
Get the day's AI news in one short email. Free.
Comments
Keep comments relevant and respectful. No links, advertising, self-promotion, or spam.