Patcher: Post-Hoc Patching of Backdoored Large Language Models

• URL: http://arxiv.org/abs/2606.02995v1
• Date: Tue, 02 Jun 2026 01:11:58 GMT
• Title: Patcher: Post-Hoc Patching of Backdoored Large Language Models
• Abstract summary: This paper presents Patcher, a post-hoc defense framework that repairs backdoored language models using only a single reported failure case and the model parameters.<n>We show that Patcher successfully localizes triggers and neutralizes backdoors while maintaining model utility.<n>This work represents a significant step toward practical defenses against training-time attacks in deployed language models.
• Score: 9.458292549037155
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Large language models remain vulnerable to jailbreak backdoor attacks, where adversaries poison safety alignment data to embed hidden triggers that bypass safety mechanisms. Existing defenses often require comprehensive attack information or multiple triggered examples, making them impractical when defenders only observe a single reported failure case without knowing whether it stems from a backdoor attack or a natural alignment bug. This paper presents Patcher, a post-hoc defense framework that repairs backdoored language models using only a single reported failure case and the model parameters. Patcher operates in two stages. First, it localizes backdoor triggers by computing response-conditioned gradient-based saliency scores and applying adaptive clustering to separate triggers from benign context. Second, it patches the model through a constrained fine-tuning objective that breaks the trigger-response association while preserving benign-task utility and robustness to non-triggered jailbreak attacks through KL-divergence constraints. We conduct extensive evaluations across multiple backdoor attack strategies and demonstrate that Patcher successfully localizes triggers and neutralizes backdoors while maintaining model utility. We further show robustness against adaptive attacks designed to evade our defense. This work represents a significant step toward practical defenses against training-time attacks in deployed language models.

Related papers

• ToxScreen: Detecting Whether an LLM Has Been Poisoned [1.2598422062985144]
  We release ToxScreen, a benchmark of roughly 800 backdoored models spanning attack objectives, trigger mechanisms, poisoning rates, model scales, and backdoor training mechanisms.<n>We find that gradient-based prompt optimization fails in recovery, whereas a token look-up that ranks candidates by attack-success rate recovers the trigger wherever the backdoor is effective.<n>No method reliably surfaces every backdoor, but a broadly jailbreakable model is itself anomalous, a useful signal even when the exact trigger is not recovered.
  arXiv  Detail & Related papers  (2026-07-29T12:35:58Z)
• Inference-Time Backdoors via Hidden Instructions in LLM Chat Templates [3.823638706744939]
  Chat templates are executable Jinja2 programs invoked at every inference call.<n>We show that an adversary who distributes a model with a maliciously modified template can implant an inference-time backdoor.<n>Backdoors generalize across inference runtimes and evade all automated security scans applied by the largest open-weight distribution platform.
  arXiv  Detail & Related papers  (2026-02-04T15:28:53Z)
• Backdoor Collapse: Eliminating Unknown Threats via Known Backdoor Aggregation in Language Models [75.29749026964154]
  Ourmethod reduces the average Attack Success Rate to 4.41% across multiple benchmarks.<n>Clean accuracy and utility are preserved within 0.5% of the original model.<n>The defense generalizes across different types of backdoors, confirming its robustness in practical deployment scenarios.
  arXiv  Detail & Related papers  (2025-10-11T15:47:35Z)
• Revisiting Backdoor Attacks on LLMs: A Stealthy and Practical Poisoning Framework via Harmless Inputs [54.90315421117162]
  We propose a novel poisoning method via completely harmless data.<n>Inspired by the causal reasoning in auto-regressive LLMs, we aim to establish robust associations between triggers and an affirmative response prefix.<n>We observe an interesting resistance phenomenon where the LLM initially appears to agree but subsequently refuses to answer.
  arXiv  Detail & Related papers  (2025-05-23T08:13:59Z)
• Jailbreaking? One Step Is Enough! [6.142918017301964]
  Large language models (LLMs) excel in various tasks but remain vulnerable to jailbreak attacks, where adversaries manipulate prompts to generate harmful outputs.<n>We propose a Reverse Embedded Defense Attack (REDA) mechanism that disguises the attack intention as the "defense" intention.<n>To enhance the model's confidence and guidance in "defensive" intentions, we adopt in-context learning (ICL) with a small number of attack examples.
  arXiv  Detail & Related papers  (2024-12-17T07:33:41Z)
• Exploiting the Vulnerability of Large Language Models via Defense-Aware Architectural Backdoor [0.24335447922683692]
  We introduce a new type of backdoor attack that conceals itself within the underlying model architecture. The add-on modules of model architecture layers can detect the presence of input trigger tokens and modify layer weights. We conduct extensive experiments to evaluate our attack methods using two model architecture settings on five different large language datasets.
  arXiv  Detail & Related papers  (2024-09-03T14:54:16Z)
• Mitigating Backdoor Attack by Injecting Proactive Defensive Backdoor [63.84477483795964]
  Data-poisoning backdoor attacks are serious security threats to machine learning models. In this paper, we focus on in-training backdoor defense, aiming to train a clean model even when the dataset may be potentially poisoned. We propose a novel defense approach called PDB (Proactive Defensive Backdoor)
  arXiv  Detail & Related papers  (2024-05-25T07:52:26Z)
• BadCLIP: Dual-Embedding Guided Backdoor Attack on Multimodal Contrastive Learning [85.2564206440109]
  This paper reveals the threats in this practical scenario that backdoor attacks can remain effective even after defenses. We introduce the emphtoolns attack, which is resistant to backdoor detection and model fine-tuning defenses.
  arXiv  Detail & Related papers  (2023-11-20T02:21:49Z)
• From Shortcuts to Triggers: Backdoor Defense with Denoised PoE [51.287157951953226]
  Language models are often at risk of diverse backdoor attacks, especially data poisoning. Existing backdoor defense methods mainly focus on backdoor attacks with explicit triggers. We propose an end-to-end ensemble-based backdoor defense framework, DPoE, to defend various backdoor attacks.
  arXiv  Detail & Related papers  (2023-05-24T08:59:25Z)
• On the Vulnerability of Backdoor Defenses for Federated Learning [8.345632941376673]
  Federated Learning (FL) is a popular distributed machine learning paradigm that enables jointly training a global model without sharing clients' data. In this paper, we study whether the current defense mechanisms truly neutralize the backdoor threats from federated learning. We propose a new federated backdoor attack method for possible countermeasures.
  arXiv  Detail & Related papers  (2023-01-19T17:02:02Z)
• Rethinking the Trigger of Backdoor Attack [83.98031510668619]
  Currently, most of existing backdoor attacks adopted the setting of emphstatic trigger, $i.e.,$ triggers across the training and testing images follow the same appearance and are located in the same area. We demonstrate that such an attack paradigm is vulnerable when the trigger in testing images is not consistent with the one used for training.
  arXiv  Detail & Related papers  (2020-04-09T17:19:37Z)
• Clean-Label Backdoor Attacks on Video Recognition Models [87.46539956587908]
  We show that image backdoor attacks are far less effective on videos. We propose the use of a universal adversarial trigger as the backdoor trigger to attack video recognition models. Our proposed backdoor attack is resistant to state-of-the-art backdoor defense/detection methods.
  arXiv  Detail & Related papers  (2020-03-06T04:51:48Z)

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.