A variational autoencoder-based model, CktGen, that directly generates analog circuits from specified requirements by mapping circuits and specifications into a joint latent space, employing contrastive learning and classifier guidance to enhance the consistency between the two modalities.
Large Language Models (LLMs) can be leveraged to automate both the design and verification of hardware modules, potentially streamlining the digital design pipeline.
A semi-automatic software prototype for generating patient-specific cranial implants by aligning healthy skull templates to the damaged target and reconstructing the defect area in voxel space.
The proposed ContrastCAD model effectively captures semantic information within the construction sequences of CAD models through contrastive learning. It also introduces a new CAD data augmentation method called Random Replace and Extrude (RRE) to enhance the learning performance of the model.
本論文は、CADプログラムの意味的コメントを自動生成するアルゴリズムとベンチマークを提案する。