ResearchBenchmarksReward prediction on Out-of-distribution task dataset (test)Reward prediction on Out-of-distribution task dataset (test)Follow0.188RMSEWEAVER0.179120.239060.2990.35894Jun 11, 2026Evaluation resultsMethodMethodLinksRMSESpearman's RhoPearson's RMMRV ScoreWEAVERFine-tuning=trueFine-tuning=true2026.060.1880.870.8630.035WEAVERFine-tuning=falseFine-tuning=false2026.060.3590.5940.5630.155CtrlWorldFine-tuning=falseFine-tuning=false2026.060.410.5230.5520.215