ResearchBenchmarksQuestion Answering on GSM8K ZHFollow9.78AccuracyBaseline-9.2208-4.28790.6455.5779May 26, 2025Evaluation ResultsMethodMethodLinksAccuracyBaselineForgetting Language=noneForgetting Language=none2025.059.78LWF-mixedForgetting Language=mixedForgetting Language=mixed2025.057.77LWFForgetting Language=ENForgetting Language=EN2025.052.35LWFForgetting Language=ITForgetting Language=IT2025.05-5.42LWFForgetting Language=ESForgetting Language=ES2025.05-6.24LWFForgetting Language=TRForgetting Language=TR2025.05-8.49