arXiv：cs.LG· Jiran Tao, Binyan Jiang·· 4 小时前AI 评分30
PairAudit：用图 token 在分布偏移下引导人工审查
PairAudit: Guiding Human Review with Graph Tokens under Distribution Shift
AI 导读
PairAudit 通过图 token 捕捉相连节点间的预测模式，利用异常关系模式发现已有预测中被忽略的错误，在固定审查预算下改进人工审查优先级。在多项安全任务实验中，PairAudit 平均比基于不确定性的审查纠正更多错误，包括更多未见攻击的错误，且计入全部审查成本、无需重新训练检测器。
正文
Abstract:Intrusion detectors can confidently misclassify attacks that were not seen during training. Human review can correct these errors, but only a limited number of cases can be checked. Uncertainty-based review may overlook confident errors, while anomaly scores alone do not show whether changing the review plan will correct more errors. We introduce PairAudit to find overlooked errors and improve review under a fixed budget. Its graph tokens capture prediction patterns across connected nodes. Rather than building another predictor through feature aggregation, PairAudit uses unusual relational patterns to uncover potential errors in existing predictions. Human feedback then helps decide whether these findings justify changing review priorities. Experiments across security tasks show that PairAudit corrects more errors on average than uncertainty-based review, including more errors on unseen attacks. These gains account for all review costs and do not require retraining the detector.
| Comments: | 22 pages, 3 figures |
| Subjects: | Machine Learning (cs.LG); Cryptography and Security (cs.CR) |
| Cite as: | arXiv:2610.10260 [cs.LG] |
| (or arXiv:2610.10260v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.10260 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Jiran Tao [view email]
[v1]
Wed, 7 Oct 2026 15:35:02 UTC (225 KB)
来源：arXiv：cs.LG · arxiv.org