[论文] Pangram 4 Technical Report

论文概要

研究领域: NLP 作者: Ben Glickenhaus, Katherine Thai, Jenna Russell, Elyas Masrour, Yue Han, Max Spero, Bradley Emi 发布时间: 2026-07-29 arXiv: 2607.27183

中文摘要

本文介绍Pangram 4，Pangram Labs最新的基于深度学习的AI文本分类模型。我们实现了0.9916的AUROC，假阳性率0.0041%，假阴性率0.3396%。与Pangram 3相比，Pangram 4不仅整体准确率更高，还表现出更优越的分布外泛化能力和对抗攻击鲁棒性。Pangram 4的另一个新颖贡献是其区分细粒度编辑和混合AI-人类合著文本的能力。我们在边界检测任务和交错式AI辅助检测方面都展示了改进。最后，我们在标准AI检测基准上的指标表明，Pangram 4在多种设置和领域下均达到了AI文本检测任务的最先进性能。

原文摘要

We present Pangram 4, the latest deep-learning-based AI-text classification model from Pangram Labs. We achieve an AUROC of 0.9916 with a false positive rate of 0.0041% and a false negative rate of 0.3396%. In addition to its increased overall accuracy compared with Pangram 3, Pangram 4 exhibits superior out-of-distribution generalization and对抗攻击robustness. Another novel contribution of Pangram 4 is its improved ability to distinguish fine-grained edits and mixed AI-human co-authored text. We demonstrate improvements to both boundary detection tasks and the detection of interleaved AI assistance. Finally, we report metrics on standard AI detection benchmarks showing that Pangram 4 achieves state-of-the-art performance on the AI text detection task across a wide variety of settings and doma...

--- *自动采集于 2026-07-31*

#论文 #arXiv #NLP #小凯