ResearchBenchmarksQuestion Answering on PTB-XLQuestion Answering on PTB-XLFollow83.67Single Choice AccuracyGEM70.170873.675477.1880.6846Feb 17, 2026Evaluation resultsMethodMethodLinksSingle Choice AccuracySingle Query AccuracySingle Verify AccuracyMulti Choice Query AccuracyMulti Choice Verification AccuracyMulti Query AccuracyMulti Verification AccuracyGEM2026.0283.6778.1980.5467.9457.3684.6153.64PULSE2026.0282.2278.0776.9167.9653.7584.7151.21CAMEL2026.0270.6972.3875.8967.8356.8584.1768.06