Each Excel workbook contains 75 objects: five objects from each of the 15 region × object type strata. Each row includes an embedded primary/front-view image immediately before the reference title.
Dataset Card
Each Excel workbook contains 75 objects: five objects from each of the 15 region × object type strata. Each row includes an embedded primary/front-view image immediately before the reference title. The two copies for a group contain the same objects, references, and model outputs. Only the annotator column differs. 每份 Excel 工作簿包含 75 件藏品，即 15 个 region × object type 类别各 5 件。每行在 reference title 前嵌入一张藏品的主要／正面图片。同一组的两份文件包含完全相同的藏品、参考标签和模型输出，仅末列 annotator 不同。 1. Annotate the five fields independently: title, culture, period, origin, and creator. 2. For each model and field, enter exactly 1 or 0 in the adjacent match0or1 column. - 1: the model output is fully or broadly consistent with at least one…
Excerpt from the card by AI4Museum.
Details
- Repository
- Carolyn-Jiang/Appear2Meaning
- Publisher
- AI4Museum
- Task category
- Visual question answering
- Tags
- Not stated by the source
- Size category
- Not stated by the source
- Languages
- en, zh
- Revision
- 531ccd91421b2acf4af10fc4fe7ac3fe03b374b9
- Last updated
- 2026-09-19
Files
22 files, 20.7 MB in total.
Every file
| File | Type | Size | SHA-256 |
|---|---|---|---|
| human_evaluation_groups/group_1.json | Data | 79.5 KB | — |
| human_evaluation_groups/group_2.json | Data | 74.6 KB | — |
| human_evaluation_groups/group_3.json | Data | 76.3 KB | — |
| llm_judge_evaluation/inputs/sampled_675_records.json | Data | 924.7 KB | — |
| llm_judge_evaluation/results/deepseek_v4_flash_judge.json | Data | 1.5 MB | — |
| llm_judge_evaluation/results/gemini_3_5_flash_lite_judge.json | Data | 1.7 MB | — |
| llm_judge_evaluation/results/gpt_4_1_mini_judge.json | Data | 1.8 MB | — |
| llm_judge_evaluation/summary/judge_comparison.json | Data | 5.8 KB | — |
| llm_judge_evaluation/summary/judge_summary.csv | Data | 3.4 KB | — |
| llm_judge_evaluation/summary/judge_summary.json | Data | 8.3 KB | — |
| reference_labels.json | Data | 727.6 KB | — |
| README.md | Documentation | 13.2 KB | — |
| llm_judge_evaluation/code/judge_prompt.md | Documentation | 2.3 KB | — |
| human_annotations/group_1_annotator_A.xlsx | Other | 2.4 MB | 6cdc5106f09b |
| human_annotations/group_1_annotator_B.xlsx | Other | 2.4 MB | 6ffdd211fdc0 |
| human_annotations/group_2_annotator_B.xlsx | Other | 2.1 MB | fcb52eff40dd |
| human_annotations/group_2_annotator_C.xlsx | Other | 2.1 MB | bf190ef1f0bd |
| human_annotations/group_3_annotator_A.xlsx | Other | 2.4 MB | a200ad262a73 |
| human_annotations/group_3_annotator_C.xlsx | Other | 2.4 MB | 185a7a571caf |
| llm_judge_evaluation/code/open_source.env.example | Other | 298 B | — |
| llm_judge_evaluation/code/run_three_judges.py | Other | 17.3 KB | — |
| .gitattributes | Repository | 3.0 KB | — |
License and Download
- License
- Not stated by the source
- Access
- No access gate
Released by AI4Museum through its official repository on Hugging Face.