nlp-waseda/VisRecall
收藏Hugging Face2025-05-22 更新2025-07-05 收录
下载链接：
https://hf-mirror.com/datasets/nlp-waseda/VisRecall
下载链接
链接失效反馈官方服务：
资源简介：
VisRecall多语言数据集，旨在评估多模态大型语言模型在跨语言视觉描述生成任务中的表现。包含450个地标在9种语言中的描述，用于测试模型在生成不同语言描述时的视觉记忆一致性。
The VisRecall multilingual dataset is designed to evaluate the performance of multimodal large language models in cross-lingual visual description generation tasks. It contains descriptions of 450 landmarks in 9 languages, used to test the consistency of visual memory in generating descriptions across different languages.
提供机构：
nlp-waseda
