ResearchBenchmarksClassification on CSNFollow12.84F1 ScoreCAMEL4.68646.80328.9211.0368Feb 17, 2026Evaluation ResultsMethodMethodLinksF1 ScoreCAMEL2026.0212.84PULSE2026.0212.62GEM2026.028.19MERL2026.026.38MELP2026.025