MLP-VLM/

bllossom-vision

	Dataset Description




	
	aihub-caption_koen

{ "image": <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=336x372>, "lang": "en", "instruction": "Summarize what is happening in the scene in a concise manner", "response": "The image appears to be a vintage photograph...", }

{ "file_id": 291040, "images": <PIL.JpegImagePlugin.JpegImageFile image mode=RGB size=1223x680>, "type": "기본표", "task": "markdown"… See the full description on the dataset page: https://huggingface.co/datasets/MLP-VLM/bllossom-vision.