AnalogRetriever: Learning Cross-Modal Representations for Analog Circuit Retrieval

TL;DR

AnalogRetriever introduces a cross-modal retrieval framework for analog circuits, enabling effective search across heterogeneous representations by learning shared embeddings, significantly improving retrieval accuracy and aiding circuit design tasks.

Contribution

It presents a novel tri-modal learning approach with a new dataset and curriculum contrastive learning, advancing cross-modal analog circuit retrieval.

Findings

Achieves 75.2% Recall@1 across retrieval tasks.

Significantly outperforms existing baselines.

Improves functional pass rates in circuit design.

Abstract

Analog circuit design relies heavily on reusing existing intellectual property (IP), yet searching across heterogeneous representations such as SPICE netlists, schematics, and functional descriptions remains challenging. Existing methods are largely limited to exact matching within a single modality, failing to capture cross-modal semantic relationships. To bridge this gap, we present AnalogRetriever, a unified tri-modal retrieval framework for analog circuit search. We first build a high-quality dataset on top of Masala-CHAI through a two-stage repair pipeline that raises the netlist compile rate from 22\% to 100\%. Built on this foundation, AnalogRetriever encodes schematics and descriptions with a vision-language model and netlists with a port-aware relational graph convolutional network, mapping all three modalities into a shared embedding space via curriculum contrastive learning. Experiments show that AnalogRetriever achieves an average Recall@1 of 75.2\% across all six cross-modal retrieval directions, significantly outperforming existing baselines. When integrated into the AnalogCoder agentic framework as a retrieval-augmented generation module, it consistently improves functional pass rates and enables previously unsolved tasks to be completed. Our code and dataset will be released.