CLIP-Inspector: Model-Level Backdoor Detection for Prompt-Tuned CLIP via OOD Trigger Inversion

• URL: http://arxiv.org/abs/2604.09101v1
• Date: Fri, 10 Apr 2026 08:33:56 GMT
• Title: CLIP-Inspector: Model-Level Backdoor Detection for Prompt-Tuned CLIP via OOD Trigger Inversion
• Abstract summary: We introduce CLIP-Inspector (CI), a backdoor detection method designed for prompt-tuned CLIP models.<n>CI reconstructs possible triggers for each class to determine if the model exhibits backdoor behaviour or not.<n>We demonstrate that using CI's reconstructed trigger for fine-tuning on correctly labeled triggered inputs enables us to re-align the model.
• Score: 9.120160208679133
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Organisations with limited data and computational resources increasingly outsource model training to Machine Learning as a Service (MLaaS) providers, who adapt vision-language models (VLMs) such as CLIP to downstream tasks via prompt tuning rather than training from scratch. This semi-honest setting creates a security risk where a malicious provider can follow the prompt-tuning protocol yet implant a backdoor, forcing triggered inputs to be classified into an attacker-chosen class, even for out-of-distribution (OOD) data. Such backdoors leave encoders untouched, making them undetectable to existing methods that focus on encoder corruption. Other data-level methods that sanitize data before training or during inference, also fail to answer the critical question, "Is the delivered model backdoored or not?" To address this model-level verification problem, we introduce CLIP-Inspector (CI), a backdoor detection method designed for prompt-tuned CLIP models. Assuming white-box access to the delivered model and a pool of unlabeled OOD images, CI reconstructs possible triggers for each class to determine if the model exhibits backdoor behaviour or not. Additionally, we demonstrate that using CI's reconstructed trigger for fine-tuning on correctly labeled triggered inputs enables us to re-align the model and reduce backdoor effectiveness. Through extensive experiments across ten datasets and four backdoor attacks, we demonstrate that CI can reconstruct effective triggers in a single epoch using only 1,000 OOD images, achieving a 94% detection accuracy (47/50 models). Compared to adapted trigger-inversion baselines, CI yields a markedly higher AUROC score (0.973 vs 0.495/0.687), thus enabling the vetting and post-hoc repair of prompt-tuned CLIP models to ensure safe deployment.

Related papers

• Self-Purification Mitigates Backdoors in Multimodal Diffusion Language Models [74.1970982768771]
  We show that well-established data-poisoning pipelines can successfully implant backdoors into MDLMs.<n>We introduce a backdoor defense framework for MDLMs named DiSP (Diffusion Self-Purification)
  arXiv  Detail & Related papers  (2026-02-24T15:47:52Z)
• Assimilation Matters: Model-level Backdoor Detection in Vision-Language Pretrained Models [71.44858461725893]
  Given a model fine-tuned by an untrusted third party, determining whether the model has been injected with a backdoor is a critical and challenging problem.<n>Existing detection methods usually rely on prior knowledge of training dataset, backdoor triggers and targets.<n>We introduce Assimilation Matters in DETection (AMDET), a novel model-level detection framework that operates without any such prior knowledge.
  arXiv  Detail & Related papers  (2025-11-29T06:20:00Z)
• Detecting Backdoor Samples in Contrastive Language Image Pretraining [32.85582585781569]
  Contrastive language-image pretraining (CLIP) has been found to be vulnerable to poisoning backdoor attacks.<n>This raises security concerns on the current practice of pretraining large-scale models on unscrutinized web data using CLIP.
  arXiv  Detail & Related papers  (2025-02-03T14:21:05Z)
• ASPIRER: Bypassing System Prompts With Permutation-based Backdoors in LLMs [17.853862145962292]
  We introduce a novel backdoor attack that systematically bypasses system prompts. Our method achieves an attack success rate (ASR) of up to 99.50% while maintaining a clean accuracy (CACC) of 98.58%.
  arXiv  Detail & Related papers  (2024-10-05T02:58:20Z)
• BackdoorIndicator: Leveraging OOD Data for Proactive Backdoor Detection in Federated Learning [7.528642177161784]
  In a federated learning (FL) system, decentralized data owners (clients) could upload their locally trained models to a central server, to jointly train a global model. Malicious clients may plant backdoors into the global model through uploading poisoned local models, causing misclassification to a target class when encountering attacker-defined triggers. Existing backdoor defenses show inconsistent performance under different system and adversarial settings, especially when the malicious updates are made statistically close to the benign ones. We propose a novel proactive backdoor detection mechanism for FL named BackdoorIndicator, which has the server inject indicator tasks into the global model leveraging out-
  arXiv  Detail & Related papers  (2024-05-31T14:44:57Z)
• Lazy Layers to Make Fine-Tuned Diffusion Models More Traceable [70.77600345240867]
  A novel arbitrary-in-arbitrary-out (AIAO) strategy makes watermarks resilient to fine-tuning-based removal. Unlike the existing methods of designing a backdoor for the input/output space of diffusion models, in our method, we propose to embed the backdoor into the feature space of sampled subpaths. Our empirical studies on the MS-COCO, AFHQ, LSUN, CUB-200, and DreamBooth datasets confirm the robustness of AIAO.
  arXiv  Detail & Related papers  (2024-05-01T12:03:39Z)
• Model Pairing Using Embedding Translation for Backdoor Attack Detection on Open-Set Classification Tasks [63.269788236474234]
  We propose to use model pairs on open-set classification tasks for detecting backdoors. We show that this score, can be an indicator for the presence of a backdoor despite models being of different architectures. This technique allows for the detection of backdoors on models designed for open-set classification tasks, which is little studied in the literature.
  arXiv  Detail & Related papers  (2024-02-28T21:29:16Z)
• BadCLIP: Trigger-Aware Prompt Learning for Backdoor Attacks on CLIP [55.33331463515103]
  BadCLIP is built on a novel and effective mechanism in backdoor attacks on CLIP. It consists of a learnable trigger applied to images and a trigger-aware context generator, such that the trigger can change text features via trigger-aware prompts.
  arXiv  Detail & Related papers  (2023-11-26T14:24:13Z)
• CleanCLIP: Mitigating Data Poisoning Attacks in Multimodal Contrastive Learning [63.72975421109622]
  CleanCLIP is a finetuning framework that weakens the learned spurious associations introduced by backdoor attacks. CleanCLIP maintains model performance on benign examples while erasing a range of backdoor attacks on multimodal contrastive learning.
  arXiv  Detail & Related papers  (2023-03-06T17:48:32Z)
• CrowdGuard: Federated Backdoor Detection in Federated Learning [39.58317527488534]
  This paper presents a novel defense mechanism, CrowdGuard, that effectively mitigates backdoor attacks in Federated Learning. CrowdGuard employs a server-located stacked clustering scheme to enhance its resilience to rogue client feedback. The evaluation results demonstrate that CrowdGuard achieves a 100% True-Positive-Rate and True-Negative-Rate across various scenarios.
  arXiv  Detail & Related papers  (2022-10-14T11:27:49Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.