ResearchBenchmarksSummarization on FeedSum (test)Follow0.022ECE (Instance)QAB-0.217121.396943.0114.62506Apr 19, 2026Evaluation ResultsMethodMethodLinksECE (Instance)GECE (Instance)Brier Score (Instance)Calibration Slope (Instance)Accuracy (Instance)ECE (Average)GECE (Average)Brier Score (Average)Calibration Slope (Average)Mean ScoreQABIndicator=GainIndicator=Gain2026.040.0220.0280.0050.147-0.0430.0330.0230.0070.6120.093GIRBIndicator=GainIndicator=Gain2026.040.0270.030.0080.1610.0170.0350.0260.0080.5940.101NoneIndicator=WinsIndicator=Wins2026.041010300000.56QABIndicator=WinsIndicator=Wins2026.041223041432.22GIRBIndicator=WinsIndicator=Wins2026.046553366634.78