Lambert 还指出了一个技术层面很少被外界提及的问题：不同模型之间存在微妙的数据分布差异。
Карина Черных (Редактор отдела «Ценности»)
。关于这个话题，safew官方下载提供了深入分析
SelectWhat's included
The website you are visiting is protected.
This does not mean confusables.txt is wrong. It means confusables.txt is a visual-similarity claim that has never been empirically validated at scale. Many entries map characters to the same abstract target under NFKC decomposition (mathematical bold A to A, for instance), and the mapping is semantically correct even if the glyphs look nothing alike. But if you treat every confusables.txt entry as equally dangerous for UI security, you are generating massive false positive rates for 96.5% of the dataset.