benchmarksfactbullishThe prediction-powered evaluation framework applies broadly to non-verifiable tasksapplies broadly to non-verifiable tasksMachine Learning (Statistics)29 Aug 2026http://arxiv.org/abs/2608.26638v1