ResearchBenchmarksQuestion Answering on HotpotQA (test) (Accuracy Metric Only)Follow14.2AccuracyMIPROv29.488810.711911.93513.1581May 13, 2026Evaluation ResultsMethodMethodLinksAccuracyMIPROv2Optimizer=MIPROv2, Bas...Optimizer=MIPROv2, Base LLM=GPT-OSS-120B2026.0514.2CANTANTEOptimizer=CANTANTE, Ba...Optimizer=CANTANTE, Base LLM=GPT-OSS-120B2026.0511.93GEPAOptimizer=GEPA, Base L...Optimizer=GEPA, Base LLM=GPT-OSS-120B2026.0510.93InitialOptimizer=None (Initia...Optimizer=None (Initial Prompts), Base LLM=GPT-OSS-120B2026.059.67