ResearchBenchmarksCriteria Alignment on ICLR submissions 2024Follow3.34Judge 1 ScoreXtraGPT2.42482.66242.93.1376May 16, 2025Evaluation ResultsMethodMethodLinksJudge 1 ScoreJudge 2 ScoreJudge 3 ScoreAggregated Alignment ScoreXtraGPT2025.053.343.74.093.71XtraGPT2025.052.463.234.013.23