Researchers have developed PRISM-UDE, a novel framework for creating accurate and physically interpretable transistor models. This approach embeds a neural network within a physics-based model to learn complex transport behaviors, which is then distilled into a closed-form symbolic expression. When applied to 3nm FinFET transistors, PRISM-UDE significantly reduced prediction error and maintained accuracy while removing the neural network, demonstrating stable performance in circuit simulations. AI
IMPACT This framework could lead to more stable and interpretable circuit simulations for advanced semiconductor nodes.
RANK_REASON Academic paper detailing a new modeling framework. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
