Diffusion Theory as a Scalpel: Detecting and Purifying Poisonous Dimensions in Pre-trained Language Models Caused by Backdoor or Bias

• URL: http://arxiv.org/abs/2305.04547v1
• Date: Mon, 8 May 2023 08:40:30 GMT
• Title: Diffusion Theory as a Scalpel: Detecting and Purifying Poisonous Dimensions in Pre-trained Language Models Caused by Backdoor or Bias
• Authors: Zhiyuan Zhang, Deli Chen, Hao Zhou, Fandong Meng, Jie Zhou, Xu Sun
• Abstract summary: Pre-trained Language Models (PLMs) may be poisonous with backdoors or bias injected by the suspicious attacker during the fine-tuning process. We propose the Fine-purifying approach, which utilizes the diffusion theory to study the dynamic process of fine-tuning for finding potentially poisonous dimensions. To the best of our knowledge, we are the first to study the dynamics guided by the diffusion theory for safety or defense purposes.
• Score: 64.81358555107788
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Pre-trained Language Models (PLMs) may be poisonous with backdoors or bias injected by the suspicious attacker during the fine-tuning process. A core challenge of purifying potentially poisonous PLMs is precisely finding poisonous dimensions. To settle this issue, we propose the Fine-purifying approach, which utilizes the diffusion theory to study the dynamic process of fine-tuning for finding potentially poisonous dimensions. According to the relationship between parameter drifts and Hessians of different dimensions, we can detect poisonous dimensions with abnormal dynamics, purify them by resetting them to clean pre-trained weights, and then fine-tune the purified weights on a small clean dataset. To the best of our knowledge, we are the first to study the dynamics guided by the diffusion theory for safety or defense purposes. Experimental results validate the effectiveness of Fine-purifying even with a small clean dataset.

Related papers

• ECLIPSE: Expunging Clean-label Indiscriminate Poisons via Sparse Diffusion Purification [29.28977815669999]
  Clean-label indiscriminate poisoning attacks add invisible perturbations to correctly labeled training images. We propose a more universally effective, practical, and robust defense scheme called ECLIPSE.
  arXiv  Detail & Related papers  (2024-06-21T12:14:24Z)
• PSBD: Prediction Shift Uncertainty Unlocks Backdoor Detection [57.571451139201855]
  Prediction Shift Backdoor Detection (PSBD) is a novel method for identifying backdoor samples in deep neural networks. PSBD is motivated by an intriguing Prediction Shift (PS) phenomenon, where poisoned models' predictions on clean data often shift away from true labels towards certain other labels. PSBD identifies backdoor training samples by computing the Prediction Shift Uncertainty (PSU), the variance in probability values when dropout layers are toggled on and off during model inference.
  arXiv  Detail & Related papers  (2024-06-09T15:31:00Z)
• Exploring Model Dynamics for Accumulative Poisoning Discovery [62.08553134316483]
  We propose a novel information measure, namely, Memorization Discrepancy, to explore the defense via the model-level information. By implicitly transferring the changes in the data manipulation to that in the model outputs, Memorization Discrepancy can discover the imperceptible poison samples. We thoroughly explore its properties and propose Discrepancy-aware Sample Correction (DSC) to defend against accumulative poisoning attacks.
  arXiv  Detail & Related papers  (2023-06-06T14:45:24Z)
• Reconstructing Graph Diffusion History from a Single Snapshot [87.20550495678907]
  We propose a novel barycenter formulation for reconstructing Diffusion history from A single SnapsHot (DASH) We prove that estimation error of diffusion parameters is unavoidable due to NP-hardness of diffusion parameter estimation. We also develop an effective solver named DIffusion hiTting Times with Optimal proposal (DITTO)
  arXiv  Detail & Related papers  (2023-06-01T09:39:32Z)
• How to Backdoor Diffusion Models? [74.43215520371506]
  This paper presents the first study on the robustness of diffusion models against backdoor attacks. We propose BadDiffusion, a novel attack framework that engineers compromised diffusion processes during model training for backdoor implantation. Our results call attention to potential risks and possible misuse of diffusion models.
  arXiv  Detail & Related papers  (2022-12-11T03:44:38Z)
• Backdoor Attacks on Pre-trained Models by Layerwise Weight Poisoning [27.391664788392]
  Pre-trained weights can be maliciously poisoned with certain triggers. Fine-tuned model will predict pre-defined labels, causing a security threat.
  arXiv  Detail & Related papers  (2021-08-31T14:47:37Z)
• Weight Poisoning Attacks on Pre-trained Models [103.19413805873585]
  We show that it is possible to construct weight poisoning'' attacks where pre-trained weights are injected with vulnerabilities that expose backdoors'' after fine-tuning. Our experiments on sentiment classification, toxicity detection, and spam detection show that this attack is widely applicable and poses a serious threat.
  arXiv  Detail & Related papers  (2020-04-14T16:51:42Z)

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.