BadEdit: Backdooring large language models by model editing

no code implementations • 20 Mar 2024 • Yanzhou Li, Tianlin Li, Kangjie Chen, Jian Zhang, Shangqing Liu, Wenhan Wang, Tianwei Zhang, Yang Liu

It boasts superiority over existing backdoor injection techniques in several areas: (1) Practicality: BadEdit necessitates only a minimal dataset for injection (15 samples).

Backdoor Attack knowledge editing

Multi-target Backdoor Attacks for Code Pre-trained Models

no code implementations • 14 Jun 2023 • Yanzhou Li, Shangqing Liu, Kangjie Chen, Xiaofei Xie, Tianwei Zhang, Yang Liu

We evaluate our approach on two code understanding tasks and three code generation tasks over seven datasets.

Code Generation Representation Learning

CommitBART: A Large Pre-trained Model for GitHub Commits

no code implementations • 17 Aug 2022 • Shangqing Liu, Yanzhou Li, Xiaofei Xie, Yang Liu

GitHub commits, which record the code changes with natural language messages for description, play a critical role for software developers to comprehend the software evolution.

Contrastive Learning Denoising

Unveiling Project-Specific Bias in Neural Code Models

no code implementations • 19 Jan 2022 • Zhiming Li, Yanzhou Li, Tianlin Li, Mengnan Du, Bozhi Wu, Yushi Cao, Junzhe Jiang, Yang Liu

We propose a Cond-Idf measurement to interpret this behavior, which quantifies the relatedness of a token with a label and its project-specificness.

Adversarial Robustness Vulnerability Detection