racineai/VDR_Quantum_Circuit_Papers

VDR_Quantum_Circuit_Papers – Overview

VDR_Quantum_Circuit_Papers is a curated dataset focused on quantum circuits and quantum gates, extracted exclusively from scientific research papers. This dataset emphasizes documents that contain circuit diagrams, matrix-based explanations, and detailed discussions of quantum operations.

Dataset Composition

This dataset was created using our open-source tool [VDR_pdf-to-parquet](https://github.com/RacineAIOS/VDR_pdf-to-parquet).

Scientific PDFs were sourced from public online sources. Each document was selected based on its focus on quantum circuits, with visual and mathematical representations. The processing pipeline extracted:

• —High-resolution images of quantum circuit diagrams
• —Accompanying textual content such as explanations, equations, and operations
• —Structured data for multimodal analysis and downstream tasks

We used Google’s Gemini 2.5 Pro model in a custom pipeline to generate diverse, expert-level questions that align with the content of each page.

Dataset Structure

Each sample in the dataset includes:

• —`id`: A unique identifier for each entry
• —`query`: A synthetic technical question generated from that page
• —`image`: A rendered image of the PDF page
• —`language`: Detected language of the extracted text

Purpose

This dataset is designed to support:

• —Training and evaluating vision-language models on technical quantum content (especially quantum circuits)
• —Multimodal document understanding and retrieval for quantum computing
• —Recognition and analysis of quantum circuits in scientific literature
• —Research in automated extraction and interpretation of circuit diagrams and related explanations

Creators

• —Yumeng YE
• —Léo APPOURCHAUX