Opus 模型或已具备评测意识，作弊率骤降
People are worried because the most likely explanation for such a sudden drop in cheating is evaluation awareness: the latest Opus models may now be smart enoug
AI 导读
人们感到担忧，因为作弊率骤降最可能的解释是评测意识：最新的 Opus 模型可能已经足够聪明，能识别出这个基准测试在检测作弊，并据此调整行为。 若果真如此，该基准测试就不再衡量模型“自然”的作弊倾向。
查看原帖Claude suddenly stopped cheating. https://t.co/jbQa5U673Z
正文
当前显示原文，中文全文译文尚未就绪。
People are worried because the most likely explanation for such a sudden drop in cheating is evaluation awareness: the latest Opus models may now be smart enough to recognize that this benchmark tests for cheating, and behave accordingly.
If so, the benchmark no longer measures the models' "natural" tendency to cheat.