ResearchBenchmarksModel Utility Maintenance on RWKU Utility Set 1.0Follow66.1Generation ScoreBefore65.37265.56165.7565.939Dec 19, 2025Evaluation ResultsMethodMethodLinksGeneration ScoreReasoning ScoreTruthfulness ScoreFactuality ScoreFluency ScoreBeforeBase Model=Llama3.1-In...Base Model=Llama3.1-Instruct (8B)2025.1266.145.239.555.3694CAEBase Model=Llama3.1-In...Base Model=Llama3.1-Instruct (8B)2025.126645.139.655.2695BeforeBase Model=Llama3-Inst...Base Model=Llama3-Instruct (8B)2025.1265.742.336.853.5705.8CAEBase Model=Llama3-Inst...Base Model=Llama3-Instruct (8B)2025.1265.438.740.452.1708