Breaking the Tuning Barrier: Zero-Hyperparameters Yield Multi-Corner Analysis Via Learned Priors

2026-03-13

AI Summary

This work addresses the prohibitive simulation cost of circuit validation across multiple process–voltage–temperature (PVT) corners, a challenge exacerbated by combinatorial explosion. Existing approaches struggle to balance automation and accuracy, while complex AI models often require laborious hyperparameter tuning. To overcome these limitations, we propose a context-learning framework built upon a large-scale pretrained foundation model that leverages data-driven regression priors in lieu of manual design, enabling cross-PVT knowledge transfer without retraining or hyperparameter adjustment. By integrating attention mechanisms with automated feature selection—reducing dimensionality from 1,152 to 48—we achieve, for the first time, high-accuracy multi-corner analysis with zero hyperparameter tuning. The method attains state-of-the-art accuracy (average mean relative error of only 0.11%) while reducing total validation cost by over an order of magnitude, thereby breaking the traditional trade-off between automation and model complexity.

Technology Category

• Machine Learning: Hardware-aware ML
• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Yield Multi-Corner Analysis validates circuits across 25+ Process-Voltage-Temperature corners, resulting in a combinatorial simulation cost of $O(K \times N)$ where $K$ denotes corners and $N$ exceeds $10^4$ samples per corner. Existing methods face a fundamental trade-off: simple models achieve automation but fail on nonlinear circuits, while advanced AI models capture complex behaviors but require hours of hyperparameter tuning per design iteration, forming the Tuning Barrier. We break this barrier by replacing engineered priors (i.e., model specifications) with learned priors from a foundation model pre-trained on millions of regression tasks. This model performs in-context learning, instantly adapting to each circuit without tuning or retraining. Its attention mechanism automatically transfers knowledge across corners by identifying shared circuit physics between operating conditions. Combined with an automated feature selector (1152D to 48D), our method matches state-of-the-art accuracy (mean MREs as low as 0.11\%) with zero tuning, reducing total validation cost by over $10\times$.

Problem

• hyperparameter tuning
• multi-corner analysis
• circuit validation
• PVT corners
• tuning barrier

Innovation

• learned priors
• zero-hyperparameters
• in-context learning
• multi-corner analysis
• foundation model