AI RESEARCH

CHASE: Competing Hypotheses for Ambiguity-Aware Selective Prediction

arXiv CS.CV

ArXi:2605.01346v1 Announce Type: new Standard selective prediction methods typically estimate uncertainty from the output of a single predictive branch. While effective for general uncertainty estimation, these approaches often struggle under partial observability, where local temporal evidence can be contradictory and standard confidence scores become misleading. We