Reproduce “A Framework for AI-Supported Mediation in Community-based Online Collaboration” 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

Soobin Cho, Mark Zachry, David W. McDonald

Researchers from the University of Washington's HCDE department introduce an information-focused framework for AI-supported mediation in online communities, emphasizing AI's need to understand discussion content, community culture, and participant history. An initial empirical study found human experts utilize 13 distinct contextual elements to comprehend online disputes, elements often missed by current AI summarization techniques.

Key insights

  • Existing AI mediation research disproportionately focuses on mediator actions while largely overlooking the crucial informational understanding required for effective intervention.
  • For AI to genuinely mediate in community contexts, it must comprehend the nuanced interplay between discussion content, community culture, and individual participant characteristics.
  • Even within the 'content' dimension, human experts utilize a complex set of 13 specific elements for dispute comprehension, which go beyond the capabilities of current AI summarization.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.