naist-nlp/ExpArt
收藏资源简介：
Explain Artworks: ExpArt旨在增强大规模视觉语言模型（LVLMs）在分析和描述艺术作品方面的能力。该数据集从英文维基百科艺术文章中提取，包含约10,000篇文章，鼓励LVLMs基于图像（有或无标题）创建深入描述。该数据集的目标是提高LVLMs在辨别和表达艺术的历史和主题细微差别方面的能力，不仅旨在提升AI对艺术的理解和批评，还寻求在人工智能与艺术史之间建立更强的联系。数据集还引入了专门的指标来评估LVLMs在艺术解释中的有效性，重点关注其对视觉和文本线索的解释。
Explain Artworks: ExpArt aims to enhance the capabilities of large-scale vision-language models (LVLMs) in analyzing and describing artworks. Extracted from English Wikipedia art articles, this dataset comprises approximately 10,000 articles, and encourages LVLMs to generate in-depth descriptions based on images with or without accompanying captions. The goal of this dataset is to improve the ability of LVLMs to discern and convey the historical and thematic nuances of art. It not only aims to advance AI's understanding and critique of art, but also seeks to foster stronger connections between artificial intelligence and art history. Additionally, the dataset introduces specialized metrics to evaluate the effectiveness of LVLMs in art interpretation, with a focus on their interpretation of visual and textual cues.
数据集卡片 for "Explain Artworks: ExpArt"
数据集描述
数据集摘要
Explain Artworks: ExpArt 旨在提升大规模视觉语言模型（LVLMs）在分析和描述艺术品方面的能力。该数据集从英语维基百科的艺术文章中提取，鼓励 LVLMs 根据带有或不带有标题的图像创建深入的描述。这一努力旨在提高 LVLMs 在辨别和阐述艺术的历史和主题细微差别方面的熟练度。Explain Artworks: ExpArt 不仅旨在提升 AI 对艺术的了解和批评，还寻求在人工智能和艺术史之间建立更紧密的联系。该数据集包含约 10,000 篇文章，并引入了专门的指标来评估 LVLMs 在艺术解释方面的有效性，重点是它们对视觉和文本线索的解释。
支持的任务和排行榜
[更多信息需要]
语言
该数据集提供英语版本。
数据集结构
以下示例展示了两种不同的训练数据集格式。第一种包含“title”字段，而第二种则不包含。
数据实例示例（带标题）
JSON { "id": "0001_T", "title": "Mona Lisa", "conversations": [ { "from": "user", "value": "<img>/images/Mona Lisa.jpg</img> Focus on Mona Lisa and explore the history." }, { "from": "assistant", "value": "Of Leonardo da Vinci’s works, the Mona Lisa is the only portrait whose authenticity...." } ] }
数据实例示例（不带标题）
JSON { "id": "0001_NT", "conversations": [ { "from": "user", "value": "<img>/images/Mona Lisa.jpg</img> Focus on this artwork and explore the history." }, { "from": "assistant", "value": "Of Leonardo da Vinci’s works, the Mona Lisa is the only portrait whose authenticity...." } ] }
数据实例
Python from datasets import load_dataset
dataset = load_dataset("naist-nlp/ExpArt")
print(dataset)
DatasetDict({
train: Dataset({
features: [id, title, conversations],
num_rows: X # Replace X with the actual number of rows in your dataset
})
})
Example of accessing a single data instance
example = dataset[train][0] print(example)
{
"id": "0001_T",
"title": "Mona Lisa",
"conversations": [
{
"from": "user",
"value": "<img src=/images/Mona Lisa.jpg></img>
Focus on Mona Lisa and explore the history."
},
{
"from": "assistant",
"value": "Of Leonardo da Vinci’s works, the Mona Lisa is the only portrait whose authenticity...."
}
]
}
