Backdoors in Neural Models of Source Code

Author:

Ramakrishnan Goutham1,Albarghouthi Aws2

Affiliation:

1. Health at Scale Corporation,San Jose,CA

2. University of Wisconsin–Madison,Madison,WI

Funder

National Science Foundation

Publisher

IEEE

Reference42 articles.

1. JBMC: A Bounded Model Checking Tool for Verifying Java Bytecode

2. Satisfiability Modulo Theories

3. Exploring backdoor poisoning attacks against malware classifiers;severi,2020

4. You autocomplete me: Poisoning vulnerabilities in neural code completion;schuster;USENIX Security Symposium,2021

5. Generating Adversarial Examples for Holding Robustness of Source Code Processing Models

Cited by 7 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Measuring Impacts of Poisoning on Model Parameters and Embeddings for Large Language Models of Code;Proceedings of the 1st ACM International Conference on AI-Powered Software;2024-07-10

2. Vulnerabilities in AI Code Generators: Exploring Targeted Data Poisoning Attacks;Proceedings of the 32nd IEEE/ACM International Conference on Program Comprehension;2024-04-15

3. Unveiling Memorization in Code Models;Proceedings of the IEEE/ACM 46th International Conference on Software Engineering;2024-04-12

4. Stealthy Backdoor Attack for Code Models;IEEE Transactions on Software Engineering;2024-04

5. An Input-Denoising-Based Defense Against Stealthy Backdoor Attacks in Large Language Models for Code;2024

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3