Research Blog
Insights, research findings, and updates from our work in autonomous AI for semiconductor analysis.
Big GCVAE: Decision-making with Adaptive Transformer Model for Failure Root Cause Analysis in Semiconductor Industry
A novel approach combining generative models with adaptive transformers to achieve 91.7% accuracy in semiconductor failure root cause analysis, demonstrating 35x faster analysis compared to traditional methods.
Leveraging Pre-trained Models for Failure Analysis Triplets Generation
Exploring the use of pre-trained language models for generating failure analysis triplets (symptom-cause-solution) in semiconductor manufacturing, leveraging transfer learning to improve knowledge extraction from limited labeled data.
GCVAE: Generalized-Controllable Variational Autoencoder
A comprehensive derivation of the Generalized-Controllable Variational AutoEncoder (GCVAE) framework, including detailed mathematical proofs and derivations from the appendix, focusing on optimal latent space representation and disentanglement.