ResearchBenchmarksSentence-level ambiguity identification on Dataset (held-out combined)Follow73.4F1 ScorePRIG58.11262.08166.0570.019Jun 3, 2026Evaluation ResultsMethodMethodLinksF1 ScorePrecisionRecallPRIG2026.0673.480.867.5GPT-5.42026.0658.753.764.7