racineai/VDR_Quantum_Circuit_Synthetic

--- license: apache-2.0 language: - en tags: - RAG - quantum - quantum circuit - synthetic - physics - DSE task_categories: - visual-document-retrieval - text-retrieval --- # VDR_Quantum_Circuit_Synthetic – Overview **VDR_Quantum_Circuit_Synthetic** is a curated multimodal dataset focused on synthetic quantum circuits. It combines generated circuit images with expert-level technical queries to support tasks such as RAG DSE, question answering, document search, and vision-language model training. --- ## Dataset Composition This dataset was created using our open-source tool **[VDR_pdf-to-parquet](https://github.com/RacineAIOS/VDR_pdf-to-parquet/tree/main)**, adapted to handle synthetic data. Quantum circuit images and descriptions were generated programmatically using **[Qiskit’s standard gate library](https://quantum.cloud.ibm.com/docs/en/api/qiskit/circuit_library#standard-gates)**. These were then used in a custom pipeline to produce technical queries aligned with each circuit using **Google’s Gemini 2.5 Pro model**. --- ## Dataset Structure Each entry in the dataset contains: - `id`: A unique identifier for the sample - `query`: A synthetic technical question generated from a quantum circuit - `image`: A visual rendering of the circuit diagram - `language`: The detected language of the query Each synthetic circuit produces 4 unique entries: a main technical query, a secondary one, a visual-based question, and a multimodal semantic query. --- ## Purpose This dataset is designed to support: - Training and evaluating vision-language models - Developing multimodal search or retrieval systems - Research in automated question generation for quantum circuits - Exploring synthetic quantum circuits as a use case in AI workflows --- ## Authors - **Yumeng Ye** - **Léo Appourchaux** ---license: Apache-2.0 许可证 language: - 英语 tags: - 检索增强生成（Retrieval-Augmented Generation，RAG） - 量子 - 量子电路 - 合成 - 物理学 - 设计空间探索（Design Space Exploration，DSE） task_categories: - 视觉文档检索 - 文本检索 --- # VDR_量子电路合成数据集 — 概述 **VDR_量子电路合成数据集**是一套经过精心甄选与整理的多模态数据集，核心聚焦于合成量子电路。该数据集将生成的电路图像与专业级技术查询相结合，可用于支撑检索增强生成（RAG）、设计空间探索（DSE）、问答、文档检索以及视觉语言模型训练等任务。 --- ## 数据集构成 本数据集基于我们开源的工具**[VDR_pdf-to-parquet](https://github.com/RacineAIOS/VDR_pdf-to-parquet/tree/main)**构建，该工具经过适配以支持合成数据处理。 研究团队通过编程方式，依托**[Qiskit标准量子门库](https://quantum.cloud.ibm.com/docs/en/api/qiskit/circuit_library#standard-gates)**生成量子电路图像与对应描述文本。随后将上述内容接入自定义流水线，借助**谷歌Gemini 2.5 Pro模型**生成与各电路匹配的技术查询内容。 --- ## 数据集结构 数据集中的每个样本条目包含以下字段： - `id`: 样本的唯一标识符 - `query`: 基于量子电路生成的合成技术问题 - `image`: 电路原理图的可视化渲染结果 - `language`: 检测到的查询文本所用语言 每个合成量子电路可生成4条独立样本，分别为主技术查询、次级技术查询、视觉导向型问题以及多模态语义查询。 --- ## 数据集用途 本数据集旨在支撑以下研究与应用场景： - 视觉语言模型的训练与评估 - 多模态搜索或检索系统的开发 - 面向量子电路的自动化问题生成研究 - 探索将合成量子电路作为人工智能工作流中的应用场景 --- ## 作者 - **叶雨萌** - **莱奥·阿普尔绍**