跳到正文
@ArtificialAnlys· @ArtificialAnlys · X·· 15 天前AI 评分32
AI 导读

Grok 4.7 (xhigh) 的 AA-Omniscience 幻觉率低于 Grok 4.6 (high):29% 对 34%。准确率基本持平,为 47% 对 48%,整体 AA-Omniscience Index 从 30 提升至 32。https://t.co/PTBLzJQqTq

正文

Grok 4.7 (xhigh) has a lower AA-Omniscience Hallucination Rate than Grok 4.6 (high): 29% versus 34%.Accuracy is broadly unchanged at 47% versus 48%, and overall AA-Omniscience Index improves from 30 to 32. https://t.co/PTBLzJQqTq

来源:@ArtificialAnlys · x.com