ResearchBenchmarksPrompt Optimization on HoverFollow52.33ScoreGEPA34.6539.2443.8348.42May 10, 2026Evaluation ResultsMethodMethodLinksScoreOptimization Budget (# Rollouts)GEPATask Model=Qwen 3 8BTask Model=Qwen 3 8B2026.0552.337,051LEVITask Model=Qwen 3 8BTask Model=Qwen 3 8B2026.05492,870MIPROv2Task Model=Qwen 3 8BTask Model=Qwen 3 8B2026.0547.33—BaselineTask Model=Qwen 3 8BTask Model=Qwen 3 8B2026.0535.33—