Skip to search formSkip to main contentSkip to account menu

Asleep at the Keyboard? Assessing the Security of GitHub Copilot’s Code Contributions

This work systematically investigates the prevalence and conditions that can cause GitHub Copilot to recommend insecure code, and explores Copilot’s performance on three distinct code generation axes—examining how it performs given diversity of weaknesses, diversity of prompts, and diversity of domains.
View on IEEE (opens in a new tab)

NNoculation: Broad Spectrum and Targeted Treatment of Backdoored DNNs

A novel two-stage defense against backdoored neural networks (BadNets) that outperforms state-of-the-art defenses NeuralCleanse and Artificial Brain Simulation that are shown to be ineffective when their restrictive assumptions are circumvented by the attacker.

Benchmarking at the Frontier of Hardware Security: Lessons from Logic Locking

This work performs a critical review of logic locking techniques in the literature, and exposes several shortcomings, and devise a community-led benchmarking exercise to address the evaluation deficiencies.
View PDF on arXiv (opens in a new tab)

Adversarial Perturbation Attacks on ML-based CAD

An adversarial retraining strategy is proposed to improve the robustness of CNN-based hotspot detection and it is shown that this strategy significantly improves robustness (by a factor of ~3) against adversarial attacks without compromising classification accuracy.
View on ACM (opens in a new tab)

An Empirical Cybersecurity Evaluation of GitHub Copilot's Code Contributions

This work systematically investigates the prevalence and conditions that can cause GitHub Copilot to recommend insecure code, and explores Copilot’s performance on three distinct code generation axes—examining how it performs given diversity of weaknesses, diversity of prompts, and diversity of domains.

Poisoning the (Data) Well in ML-Based CAD: A Case Study of Hiding Lithographic Hotspots

This work shows that training data poisoning attacks are feasible and stealthy, demonstrating a backdoored neural network that performs normally on clean inputs but misbehaves on inputs when a backdoor trigger is present, and raises some fundamental questions about the robustness of ML-based systems in CAD.
View on IEEE (opens in a new tab)

Challenges and New Directions for AI and Hardware Security

The growing overlap betweenAI/ML and hardware for security is examined, where AI/ML techniques provide practitioners with new ways to monitor runtime behavior but also provide new tools for attackers to steal secret information.
View on IEEE (opens in a new tab)

Not All Fabrics Are Created Equal: Exploring eFPGA Parameters For IP Redaction

The results encourage designers to work with custom eFPGA fabrics rather than off-the-shelf commercial FPGAs and reveals that only considering a redaction fabric’s bitstream size is inadequate for gauging security.
View PDF on arXiv (opens in a new tab)

DAVE: Deriving Automatically Verilog from English

The use of state-of-the-art machine learning (ML) is explored to automatically derive Verilog snippets from English via fine-tuning GPT-2, a natural language ML system.
View on ACM (opens in a new tab)

OpenABC-D: A Large-Scale Dataset For Machine Learning Guided Integrated Circuit Synthesis

OpenABC-D is described, a large-scale, labeled dataset produced by synthesizing open source designs with a leading open-source logic synthesis tool and illustrated its use in developing, evaluating and benchmarking ML-guided logic synthesis.
View PDF on arXiv (opens in a new tab)
By clicking accept or continuing to use the site, you agree to the terms outlined in our Privacy Policy (opens in a new tab), Terms of Service (opens in a new tab), and Dataset License (opens in a new tab)