Securing Code Understanding: Detecting Natural Backdoor Vulnerability in Code Language Models
- 1 January 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Software Engineering
- Vol. 52 (10) , 1-18
Abstract
No abstract availableFunding Information
- National Natural Science Foundation of China (U24A20337)
- National Key Research and Development Program of China (2024YFF0908000)
- National Research Foundation, Singapore, and the Cyber Security Agency under its National Cybersecurity R&D Programme (NCRP25-P04-TAICeN)
This publication has 33 references indexed in Scilit:
- Source Code Summarization in the Era of Large Language Models2025
- Show Me Your Code! Kill Code Poisoning: A Lightweight Method Based on Code Naturalness2025
- TrojanPuzzle: Covertly Poisoning Code-Suggestion Models2024
- Distilled GPT for source code summarizationAutomated Software Engineering, 2024
- Gamma: Revisiting Template-Based Automated Program Repair Via Mask Prediction2023
- Backdoors in Neural Models of Source Code2022
- Piccolo: Exposing Complex Backdoors in NLP Transformer Models2022
- Shortcut learning in deep neural networksNature Machine Intelligence, 2020
- Universal Adversarial Perturbations2017
- Modeling and Discovering Vulnerabilities with Code Property Graphs2014