{
 "id": "xzz789/umm-sft-data",
 "owner": "xzz789",
 "name": "umm-sft-data",
 "downloads": 441,
 "likes": 1,
 "last_modified": "2026-08-24T05:10:36.000Z",
 "stated_licence": "other",
 "stated_from": "tag",
 "licence_name": "mixed-web-sources-and-odc-by-metadata",
 "licence_link": null,
 "file_licence": null,
 "licence_files": [],
 "sources": [],
 "gated": false,
 "tasks": [
  "visual-question-answering",
  "text-to-image"
 ],
 "citation": "@article{han2026towards,\n  title={Towards Physics of Multimodal Pretraining: Knowledge Flow, Modality Synergy, Early Unification, and Recipes},\n  author={Han, Junlin and Tong, Shengbang and Fan, David and Chen, Minghao and Torr, Philip and Kokkinos, Filippos and Lewis, Mike},\n  journal={arXiv preprint arXiv:2608.05000},\n  year={2026}\n}",
 "signal": "unclear",
 "reason": "The checks could not identify a recognised licence ID in the dataset's metadata.",
 "revision": "8d44c7cfc6134eca2d1a0930a329eb5503be2fe8",
 "stated_evidence": "https://huggingface.co/datasets/xzz789/umm-sft-data/blob/8d44c7cfc6134eca2d1a0930a329eb5503be2fe8/README.md",
 "file_evidence": "https://huggingface.co/datasets/xzz789/umm-sft-data/tree/8d44c7cfc6134eca2d1a0930a329eb5503be2fe8",
 "card_nc": false
}