SALAD: Systematic Assessment of Machine Unlearning on LLM-Aided Hardware Design
This addresses security concerns for hardware designers using LLMs, though it appears incremental as it applies existing unlearning techniques to a new domain.
The authors tackled data security risks in LLM-aided hardware design, such as IP leakage and malicious code, by introducing SALAD, a framework that uses machine unlearning to selectively remove contaminated or sensitive data from pre-trained models without full retraining.
Large Language Models (LLMs) offer transformative capabilities for hardware design automation, particularly in Verilog code generation. However, they also pose significant data security challenges, including Verilog evaluation data contamination, intellectual property (IP) design leakage, and the risk of malicious Verilog generation. We introduce SALAD, a comprehensive assessment that leverages machine unlearning to mitigate these threats. Our approach enables the selective removal of contaminated benchmarks, sensitive IP and design artifacts, or malicious code patterns from pre-trained LLMs, all without requiring full retraining. Through detailed case studies, we demonstrate how machine unlearning techniques effectively reduce data security risks in LLM-aided hardware design.