Language Models in Security
Table of contents
Papers
2015
- Exploiting and Protecting Dynamic Code Generation NDSS 2015
- Readactor: Practical Code Randomization Resilient to Memory Disclosure OAKLAND 2015
2016
2017
- Automated Crowdturfing Attacks and Defenses in Online Review Systems CCS 2017
- POSTER: Vulnerability Discovery with Function Representation Learning from Unlabeled Projects CCS 2017
- A4NT: Author Attribute Anonymity by Adversarial Training of Neural Machine Translation USENIX 2017
2018
- Neural Machine Translation Inspired Binary Code Similarity Comparison beyond Function Pairs NDSS 2018
- AI2: Safety and Robustness Certification of Neural Networks with Abstract Interpretation OAKLAND 2018
2019
- Analyzing Information Leakage of Updates to Natural Language Models CCS 2019
- CodeAlchemist: Semantics-Aware Code Generation to Find Vulnerabilities in JavaScript Engines NDSS 2019
- Updates-Leak: Data Set Inference and Reconstruction Attacks in Online Learning USENIX 2019
2020
- Privacy Risks of General-Purpose Language Models OAKLAND 2020
- Adversarial Watermarking Transformer: Towards Tracing Text Provenance with Data Hiding OAKLAND 2020
- TextShield: Robust Text Classification Based on Multimodal Embedding and Neural Machine Translation USENIX 2020
- Montage: A Neural Network Language Model-Guided JavaScript Engine Fuzzer USENIX 2020
2021
- Structured Leakage and Applications to Cryptographic Constant-Time and Cost CCS 2021
- Cert-RNN: Towards Certifying the Robustness of Recurrent Neural Networks CCS 2021
- Hidden Backdoors in Human-Centric Language Models CCS 2021
- Backdoor Pre-trained Models Can Transfer to All CCS 2021
- PalmTree: Learning an Assembly Language Model for Instruction Embedding CCS 2021
- Bolt-Dumbo Transformer: Asynchronous Consensus As Fast As the Pipelined BFT CCS 2021
- Get a Model! Model Hijacking Attack Against Machine Learning Models NDSS 2021
- SiRnn: A Math Library for Secure RNN Inference OAKLAND 2021
- Asleep at the Keyboard? Assessing the Security of GitHub Copilot’s Code Contributions OAKLAND 2021
- Spinning Language Models: Risks of Propaganda-As-A-Service and Countermeasures OAKLAND 2021
- Examining Zero-Shot Vulnerability Repair with Large Language Models OAKLAND 2021
- SmarTest: Effectively Hunting Vulnerable Transaction Sequences in Smart Contracts through Language Model-Guided Symbolic Execution USENIX 2021
- An Empirical Study of Training Self-Supervised Vision Transformers USENIX 2021
2022
- Order-Disorder: Imitation Adversarial Attacks for Black-box Neural Ranking Models CCS 2022
- LoneNeuron: A Highly-Effective Feature-Domain Neural Trojan Using Invisible and Polymorphic Watermarks CCS 2022
- Poster Towards Authorship Obfuscation with Language Models CCS 2022
- A Generic Methodology for the Modular Verification of Security Protocol Implementations CCS 2022
- Interpretable Federated Transformer Log Learning for Cloud Threat Forensics NDSS 2022
- Reconstructing Training Data with Informed Adversaries OAKLAND 2022
- Piccolo: Exposing Complex Backdoors in NLP Transformer Models OAKLAND 2022
- for Prediction City Region Re-Weighting USENIX 2022
- Membership Inference Attacks and Defenses in Neural Network Pruning USENIX 2022
- Lost at C: A User Study on the Security Implications of Large Language Model Code Assistants USENIX 2022
2023
- Transformer-based Model for Multi-tab Website Fingerprinting Attack CCS 2023
- Stealing the Decoding Algorithms of Language Models CCS 2023
- Large Language Models for Code: Security Hardening and Adversarial Testing CCS 2023
- Protecting Intellectual Property of Large Language Model-Based Code Generation APIs via Watermarks CCS 2023
- SalsaPicante: A Machine Learning Attack on LWE with Binary Secrets CCS 2023
- DP-Forward: Fine-tuning and Inference on Language Models with Differential Privacy in Forward Pass CCS 2023
- Read Between the Lines: Detecting Tracking JavaScript with Bytecode Classification CCS 2023
- Poster: Boosting Adversarial Robustness by Adversarial Pre-training CCS 2023
- Demo: Certified Robustness on Toolformer CCS 2023
- BARS: Local Robustness Certification for Deep Learning based Traffic Analysis Systems NDSS 2023
- Redeem Myself: Purifying Backdoors in Deep Learning Models using Self Attention Distillation OAKLAND 2023
- Robust Multi-tab Website Fingerprinting Attacks in the Wild OAKLAND 2023
- Improving Real-world Password Guessing Attacks via Bi-directional Transformers USENIX 2023
- CodexLeaks: Privacy Leaks from Code Generation Language Models in GitHub Copilot USENIX 2023
- Two-in-One: A Model Hijacking Attack Against Text Generation Models USENIX 2023
- PELICAN: Exploiting Backdoors of Naturally Trained Deep Learning Models In Binary Code Analysis USENIX 2023
- A Data-free Backdoor Injection Approach in Neural Networks USENIX 2023
- Network Detection of Interactive SSH Impostors Using Deep Learning USENIX 2023
Top Researchers in LLMSec
- Shouling Ji
- Jinfeng Li
- Ting Wang
- Ahmed Salem
- Michael Backes
- Yang Zhang
- Mario Fritz
- Hammond Pearce
- Brendan Dolan-Gavitt
- R. Karri
- Tianyu Du
- Lujia Shen
- Jie Shi
- Chengfang Fang
- Jianwei Yin
- Martin T. Vechev
- HyungSeok Han
- Sang Kil Cha
- Min Yang
- Baleegh Ahmad
- Benjamin Tan
- Yingqi Liu
- Guangyu Shen
- Guanhong Tao
- Shengwei An