ResearchBenchmarksMultiple-choice Question Answering on SST-2Follow91.38AccuracyITI88.5289.262590.00590.7475Dec 21, 2025Evaluation ResultsMethodMethodLinksAccuracyITITarget LLM=LLaMA2-7B-CHATTarget LLM=LLaMA2-7B-CHAT2025.1291.38LLM-CASTarget LLM=LLaMA2-7B-CHATTarget LLM=LLaMA2-7B-CHAT2025.1291.3CAATarget LLM=LLaMA2-7B-CHATTarget LLM=LLaMA2-7B-CHAT2025.1291.16SADITarget LLM=LLaMA2-7B-CHATTarget LLM=LLaMA2-7B-CHAT2025.1288.69BaselineTarget LLM=LLaMA2-7B-CHATTarget LLM=LLaMA2-7B-CHAT2025.1288.63