Reproduction Package for "Black-Box Adversarial Attacks on LLM-Based Code Completion" [ICML 2025]
-
Updated
Jun 16, 2025 - Python
#
adverserial-attack
Here are 4 public repositories matching this topic...
FGSM (Fast Gradient Sign Method) is an adversarial attack technique that adds small, calculated perturbations to input data to fool CNNs. Proposed by Ian Goodfellow in 2014, it generates adversarial examples to mislead the model's predictions.
-
Updated
Aug 19, 2024 - Jupyter Notebook
Auditing a content moderation model using DistilBERT on the Jigsaw dataset. Covers bias analysis, adversarial attacks (character evasion and label poisoning), mitigation techniques, and a guardrail pipeline to improve fairness, robustness, and real-world reliability.
-
Updated
Apr 21, 2026 - Jupyter Notebook
An empirical investigation into the robustness-efficiency tradeoff of PEFT methods against jailbreak attacks
-
Updated
May 4, 2026 - Python
Improve this page
Add a description, image, and links to the adverserial-attack topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the adverserial-attack topic, visit your repo's landing page and select "manage topics."