Beyond Native Success: Auditing Deployment-Interface Exposure of CLIP Backdoors 文章

ArXiv CS.CL2026-06-17NEWSen作者: Kunlan Xiang, Haomiao Yang, Wenbo Jiang

详细信息

来源站点
ArXiv CS.CL
作者
Kunlan Xiang, Haomiao Yang, Wenbo Jiang
文章类型
NEWS
语言
en
发布日期
2026-06-17

摘要

arXiv:2606.17815v1 Announce Type: cross Abstract: Contrastive Language-Image Pre-training models are widely reused across downstream interfaces, including feature extraction, retrieval, reranking, and selection. Existing CLIP backdoor, however, usually validate attacks on a small attack-native task, leaving unclear whether the same poisoned checkpoint remains exposed, weakens, or becomes not applicable when reused through other interfaces. We introduce DIFE, a Deployment-Interface Footprint Evaluation framework that audits backdoored CLIP checkpoints across deployment interfaces. DIFE makes various evaluations comparable by specifying each interface's component readout, trigger channel, target event, reference condition, and metric. DIFE also introduces effective-footprint diagnosis to identify the reusable CLIP component or component combination that carries exposure and explains where risk transfers.

相关事件

暂无数据

相关公司

暂无数据

相关人物

暂无数据