ResearchBenchmarksJob Location on Hiring Assistant val batch seed 42 (test)Job Location on Hiring Assistant val batch seed 42 (test)Follow0.594Mean KappaSPEAR-0.01440.143550.30150.45945May 25, 2026Evaluation resultsMethodMethodLinksMean KappaStd Dev KappaMean Deviation GapSPEAROptimizer=GPT-5.4Optimizer=GPT-5.42026.050.59400.25GEPAOptimizer=GPT-5.4Optimizer=GPT-5.42026.050.1730.040.28TextGradOptimizer=GPT-5.4Optimizer=GPT-5.42026.050.0090.030.21