ResearchBenchmarksAbility Estimation on Golden (test)Ability Estimation on Golden (test)Follow0.0961MAEAgentCAT0.0884880.1398690.191250.242631Jun 20, 2026Evaluation resultsMethodMethodLinksMAERMSECorrelation CoefficientAgentCAT2026.060.09610.12810.8689MFI2026.060.24060.27380.7572KLI2026.060.25730.28960.8046MAAT2026.060.27490.30220.7454Random2026.060.28640.30150.7535