AI RESEARCH
Causal Discovery as Dialectical Aggregation: A Quantitative Argumentation Framework
arXiv CS.AI
•
ArXi:2604.23633v1 Announce Type: new Constraint-based causal discovery is brittle in finite-sample regimes because erroneous conditional-independence (CI) decisions can cascade into substantial structural errors. We propose Quantitative Argumentation for Causal Discovery (QACD), a semantics-driven framework that represents CI outcomes as graded, defeasible arguments rather than irreversible constraints.