ResearchBenchmarksUtility Evaluation on LM Utility Evaluation DatasetFollow9.12Utility ScoreCB7.6128.00358.3958.7865Apr 17, 2026Evaluation ResultsMethodMethodLinksUtility ScoreCBModel=Mistral-SmallModel=Mistral-Small2026.049.12BaseModel=Mistral-SmallModel=Mistral-Small2026.048.9BaseModel=Mistral-NemoModel=Mistral-Nemo2026.048.78CBModel=Mistral-NemoModel=Mistral-Nemo2026.048.72GoalModel=Mistral-SmallModel=Mistral-Small2026.048.69GoalModel=Mistral-NemoModel=Mistral-Nemo2026.048.39BeamModel=Mistral-SmallModel=Mistral-Small2026.048.06BeamModel=Mistral-NemoModel=Mistral-Nemo2026.047.67