List your company

Multiple-choice question answering (MCQA) with evidence on RETINA-SAFE (test)

58.91E-AlignQwen2.5-7B
23.643632.799341.95551.1107Apr 7, 2026

Evaluation results

MethodLinks
Qwen2.5-7BEvaluation setting=black-box QA2026.0458.9188.1221.1456.06
Llama 2-13BEvaluation setting=black-box QA2026.0451.2416.429.1225.59
Llama 3-8BEvaluation setting=black-box QA2026.0444.1222.1512.4126.23
Llama 2-7BEvaluation setting=black-box QA2026.0438.4214.128.1520.23
Random-choiceEvaluation setting=4-way; lower bound2026.0425252525