arXiv:2510. 09260v2 Announce Type: replace-cross Abstract: Recent work has shown that RLHF is highly susceptible to backdoor attacks.
arXiv:2606. 07963v1 Announce Type: new Abstract: Backdoor attacks in large language models (LLMs) are often treated as isolated trigger-response failures, motivating defenses tailored to specific triggers or behaviors.
arXiv:2511.14301v4 Announce Type: replace-cross Abstract: Transformer-based models are highly susceptible to backdoor attacks via supervised fine-tuning (SFT). To red-team existing data-poisoning def...
UniGuardian is a training‑free detector for large language models that jointly identifies prompt injection, backdoor, and adversarial attacks—collectively called Prompt Trigger Attacks (PTA). It measures how structured prompt perturbations shift the model’s output distribution and uses a single‑forward strategy to detect attacks while generating text in a shared batched forward pass. Experiments show that UniGuardian accurately and efficiently identifies trigger‑activated prompts in LLMs.
arXiv:2610.03124v1 Announce Type: cross Abstract: Open-weight language models can be downloaded, modified, and deployed beyond their developers' control, limiting the effectiveness of centrally enfor...
arXiv:2607. 26099v1 Announce Type: cross Abstract: Machine-learning services increasingly rely on public data, third-party providers, and outsourced training, creating opportunities for data-poisoning attacks that implant persistent malicious behavior while preserving benign utility.
arXiv:2604. 05809v2 Announce Type: replace-cross Abstract: This paper presents Text-Guided Backdoor (TGB), an adjustable backdoor attack against multimodal pretrained models that uses natural-word triggers, namely words that can naturally occur in ordinary textual inputs.
The paper introduces the Cancerous Backdoor Attack, a novel backdoor technique for large language models that activates gradually through multiple hidden stages rather than a single trigger-response. It uses invisible tokens to embed intermediate stages into the model’s reasoning process and a stage controller to monitor the attack’s progression during inference. Experiments on various reasoning datasets show that this method consistently corrupts model outputs over time and outperforms existing backdoor baselines.
FedGraph is a graph-based aggregation framework designed to defend federated large language model fine-tuning against backdoor attacks. It exploits the observation that malicious clients produce consistently similar updates, while benign clients generate diverse ones, by modeling client updates as nodes in a dynamically constructed graph and extracting structural features. The resulting compact client representations enable effective identification and suppression of malicious contributions, achieving high detection accuracy and low false positives in extensive experiments.
As large language models (LLMs) are deployed in high-stakes domains, adversaries may poison training data to implant backdoors: hidden triggers that covertly manipulate model behavior at inference time. We ask whether a defender can recover such a trigger under realistic affordances, namely white-box access to the weights and knowledge of the behavior of concern, but no training data, no trusted reference model, no knowledge of the trigger, and no certainty that the model is poisoned.
Backdoor attacks in Large Language Models (LLMs) are a growing security concern, where models can generate adversary-chosen content. Existing defenses target backdoors one at a time and typically require knowledge of the trigger, leaving the defender at a structural disadvantage when unknown backdoors may exist in a model.
Parameter-Efficient Fine-tuned (PEFT) models are frequently downloaded from open repositories by practitioners. This widespread practice creates a significant attack surface, as malicious actors can publish backdoored models that induce specific behaviors in response to predefined triggers.
The paper introduces Trigger‑Neutralized Counterfactual Scoring (TNCS), a method for selecting training samples to poison pre‑trained code models with minimal data. TNCS compares model outputs on triggered inputs to outputs on trigger‑neutralized counterfactuals, enabling effective backdoor attacks on both classification and generation tasks. Experiments on CodeXGLUE tasks show higher attack success rates at very low poisoning ratios, with no degradation of clean‑task performance and transferability across trigger families.