#AI text detection

15 stories · 1 in 7 days · 4 in 30 days
Stories, products and papers tagged "AI text detection", newest first.
Nov 10
Chinese content研究:短文本 AI 检测器误报率随长度骤变,提出分长度校准方法

做内容审核或学术诚信检查的会想看这篇:它用公式证明检测器在 10 token 短文本上误报率会从 1% 飙到 29%,Binoculars 真阳性率最高也只有 59%,还给了分长度校准和自动弃权的具体做法,可以直接照着实现。

arXiv cs.AI@Ritesh Janga, Justin ZhanSource
Oct 28
Oct 23
PapersBuildersSelected04:10
Google DeepMind paper makes AI-text detector decisions interpretable

Interpretability here does not cost accuracy — on out-of-distribution text the transparent version actually beat the black box by 17 F1 points. Teams flagging student essays or peer reviews can audit which cues drive verdicts and remove ones they consider unfair, such as styles correlated with non-native writing. The method also offers a template for making other safety classifiers inspectable.

rohanpaul_ai@rohanpaul_aiSource
Oct 11
Aug 28
Aug 13
Aug 4
Aug 3
Jul 30
Jul 19
Jun 18
Jun 5
May 26
May 22