Patronus: Identifying and Mitigating Transferable Backdoors in Pre-trained Language Models

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Zhao, Tianhang, Du, Wei, Zhao, Haodong, Duan, Sufeng, Liu, Gongshen
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914185683664896
author Zhao, Tianhang
Du, Wei
Zhao, Haodong
Duan, Sufeng
Liu, Gongshen
author_facet Zhao, Tianhang
Du, Wei
Zhao, Haodong
Duan, Sufeng
Liu, Gongshen
contents Transferable backdoors pose a severe threat to the Pre-trained Language Models (PLMs) supply chain, yet defensive research remains nascent, primarily relying on detecting anomalies in the output feature space. We identify a critical flaw that fine-tuning on downstream tasks inevitably modifies model parameters, shifting the output distribution and rendering pre-computed defense ineffective. To address this, we propose Patronus, a novel framework that use input-side invariance of triggers against parameter shifts. To overcome the convergence challenges of discrete text optimization, Patronus introduces a multi-trigger contrastive search algorithm that effectively bridges gradient-based optimization with contrastive learning objectives. Furthermore, we employ a dual-stage mitigation strategy combining real-time input monitoring with model purification via adversarial training. Extensive experiments across 15 PLMs and 10 tasks demonstrate that Patronus achieves $\geq98.7\%$ backdoor detection recall and reduce attack success rates to clean settings, significantly outperforming all state-of-the-art baselines in all settings. Code is available at https://github.com/zth855/Patronus.
format Preprint
id arxiv_https___arxiv_org_abs_2512_06899
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Patronus: Identifying and Mitigating Transferable Backdoors in Pre-trained Language Models
Zhao, Tianhang
Du, Wei
Zhao, Haodong
Duan, Sufeng
Liu, Gongshen
Cryptography and Security
Transferable backdoors pose a severe threat to the Pre-trained Language Models (PLMs) supply chain, yet defensive research remains nascent, primarily relying on detecting anomalies in the output feature space. We identify a critical flaw that fine-tuning on downstream tasks inevitably modifies model parameters, shifting the output distribution and rendering pre-computed defense ineffective. To address this, we propose Patronus, a novel framework that use input-side invariance of triggers against parameter shifts. To overcome the convergence challenges of discrete text optimization, Patronus introduces a multi-trigger contrastive search algorithm that effectively bridges gradient-based optimization with contrastive learning objectives. Furthermore, we employ a dual-stage mitigation strategy combining real-time input monitoring with model purification via adversarial training. Extensive experiments across 15 PLMs and 10 tasks demonstrate that Patronus achieves $\geq98.7\%$ backdoor detection recall and reduce attack success rates to clean settings, significantly outperforming all state-of-the-art baselines in all settings. Code is available at https://github.com/zth855/Patronus.
title Patronus: Identifying and Mitigating Transferable Backdoors in Pre-trained Language Models
topic Cryptography and Security
url https://arxiv.org/abs/2512.06899