dataset card

README.md

@@ -5,11 +5,10 @@ language:
 license: cc-by-4.0
 task_categories:
 - visual-question-answering
-- image-text-to-text
 task_ids:
-- open-ended-vqa
-- multi-choice-vqa
-- visual-grounding
+- visual-question-answering
+- multiple-choice-qa
+- document-question-answering
 size_categories:
 - 10K<n<100K
 tags:
@@ -17,104 +16,176 @@ tags:
 - cultural-heritage
 - multimodal
 - benchmark
+configs:
+- config_name: gallery
+  default: true
+  data_files:
+  - split: train
+    path: viewer/gallery.jsonl
+- config_name: qa_index
+  data_files:
+  - split: train
+    path: viewer/qa_index.jsonl
+- config_name: qa_CC
+  data_files:
+  - split: train
+    path: viewer/qa_by_type/CC.jsonl
+- config_name: qa_ER_choice
+  data_files:
+  - split: train
+    path: viewer/qa_by_type/ER_choice.jsonl
+- config_name: qa_ER_fillin
+  data_files:
+  - split: train
+    path: viewer/qa_by_type/ER_fillin.jsonl
+- config_name: qa_IR
+  data_files:
+  - split: train
+    path: viewer/qa_by_type/IR.jsonl
+- config_name: qa_ITT_MHQA_choice
+  data_files:
+  - split: train
+    path: viewer/qa_by_type/ITT_MHQA_choice.jsonl
+- config_name: qa_ITT_MHQA_fillin
+  data_files:
+  - split: train
+    path: viewer/qa_by_type/ITT_MHQA_fillin.jsonl
+- config_name: qa_MITT_MHQA_choice
+  data_files:
+  - split: train
+    path: viewer/qa_by_type/MITT_MHQA_choice.jsonl
+- config_name: qa_MITT_MHQA_fillin
+  data_files:
+  - split: train
+    path: viewer/qa_by_type/MITT_MHQA_fillin.jsonl
+- config_name: qa_PR
+  data_files:
+  - split: train
+    path: viewer/qa_by_type/PR.jsonl
+- config_name: qa_SR
+  data_files:
+  - split: train
+    path: viewer/qa_by_type/SR.jsonl
+- config_name: qa_TR_choice
+  data_files:
+  - split: train
+    path: viewer/qa_by_type/TR_choice.jsonl
+- config_name: qa_TR_fillin
+  data_files:
+  - split: train
+    path: viewer/qa_by_type/TR_fillin.jsonl
+- config_name: qa_TTI
+  data_files:
+  - split: train
+    path: viewer/qa_by_type/TTI.jsonl
+- config_name: qa_VA
+  data_files:
+  - split: train
+    path: viewer/qa_by_type/VA.jsonl
 ---
+
 # KnowCP
 
 KnowCP is a benchmark for Chinese painting understanding and reasoning.
-It supports both recognition-style tasks (region/text extraction) and knowledge/reasoning tasks (open QA, multiple choice, and multi-turn QA).
+It supports recognition-style tasks (region/text extraction) and knowledge/reasoning tasks (open QA, multiple choice, and multi-turn QA).
 
 ## Dataset Summary
 
 - Name: KnowCP
 - Modality: image + text
-- Language: Chinese (Simplified/Traditional text may appear in source inscriptions)
+- Language: Chinese
 - License: CC BY 4.0
 - Total images: 2331
-- Total question items (`questions/questions_all.jsonl`): 26137
-
-## Supported Tasks
-
-The release includes multiple task files in `questions/by_type/`:
-
-- `CC.jsonl` (922)
-- `ER_choice.jsonl` (2875)
-- `ER_fillin.jsonl` (1279)
-- `IR.jsonl` (2351)
-- `ITT_MHQA_choice.jsonl` (5290)
-- `ITT_MHQA_fillin.jsonl` (5290)
-- `MITT_MHQA_choice.jsonl` (330)
-- `MITT_MHQA_fillin.jsonl` (330)
-- `PR.jsonl` (922)
-- `SR.jsonl` (1792)
-- `TR_choice.jsonl` (1312)
-- `TR_fillin.jsonl` (1312)
-- `TTI.jsonl` (1210)
-- `VA.jsonl` (922)
+- Total question items (full set): 26137
+
+## Dataset Viewer Layout
+
+This dataset has two viewer styles:
+
+- `gallery` (default): one row per image page with selected metadata.
+- `qa_*`: QA preview grouped by task type, so each type can be browsed separately.
+
+Viewer source files:
+
+- `viewer/gallery.jsonl`
+- `viewer/qa_index.jsonl`
+- `viewer/qa_by_type/*.jsonl`
+
+`qa_*` files are lightweight previews (top rows per type) for stable Hugging Face Dataset Viewer rendering.
+The full benchmark data remains in `questions/by_type/*.jsonl`.
+
+## Supported Task Files (Full Benchmark)
+
+- `questions/by_type/CC.jsonl` (922)
+- `questions/by_type/ER_choice.jsonl` (2875)
+- `questions/by_type/ER_fillin.jsonl` (1279)
+- `questions/by_type/IR.jsonl` (2351)
+- `questions/by_type/ITT_MHQA_choice.jsonl` (5290)
+- `questions/by_type/ITT_MHQA_fillin.jsonl` (5290)
+- `questions/by_type/MITT_MHQA_choice.jsonl` (330)
+- `questions/by_type/MITT_MHQA_fillin.jsonl` (330)
+- `questions/by_type/PR.jsonl` (922)
+- `questions/by_type/SR.jsonl` (1792)
+- `questions/by_type/TR_choice.jsonl` (1312)
+- `questions/by_type/TR_fillin.jsonl` (1312)
+- `questions/by_type/TTI.jsonl` (1210)
+- `questions/by_type/VA.jsonl` (922)
 
 ## Repository Structure
 
-- `questions/by_type/*.jsonl`: per-task split files
-- `images/**`: image assets used by questions
-- `kb/knowledge_base.json`: structured knowledge base
-- `annotations/**`: annotation resources
-
-## Data Format
-
-Each line in question files is a JSON object. Typical fields include:
-
-- `qid`: unique question id
-- `image_id`: image identifier
-- `phase`: benchmark phase
-- `question_no`: question index in phase
-- `type`: task type
-- `answer_format`: expected output format (`text` or `json`)
-- `prompt`: full question prompt with output constraints
-- `ground_truth`: reference answer (string or structured object)
-- `image_paths`: related image path list (relative paths)
-
-Example (single line from `questions/questions_all.jsonl`):
-
-```json
-{
-	"qid": "IMG000002-P4-Q3B",
-	"image_id": "IMG000002",
-	"phase": "P4",
-	"question_no": "Q3B",
-	"type": "CC",
-	"answer_format": "text",
-	"prompt": "请结合题材来源与图文对应关系，解释这幅作品的文化含义。",
-	"ground_truth": "...",
-	"image_paths": ["images/IMG000002_0.jpg", "images/IMG000002_1.jpg"]
-}
-```
+- `images/**`: image assets
+- `kb/knowledge_base.json`: artwork metadata (`title`, `artist`, `dynasty`, `institution`)
+- `questions/questions_all.jsonl`: merged full question file
+- `questions/by_type/*.jsonl`: full by-type benchmark files
+- `viewer/gallery.jsonl`: gallery view data
+- `viewer/qa_by_type/*.jsonl`: by-type QA preview files
+
+## Data Fields in Viewer
+
+### `gallery`
+
+- `image`
+- `image_id`
+- `page_index`
+- `title`
+- `title_en`
+- `artist`
+- `dynasty`
+- `institution`
+- `image_count`
+
+### `qa_*`
+
+- `source_file`
+- `qid`
+- `type`
+- `phase`
+- `question_no`
+- `answer_format`
+- `prompt`
+- `ground_truth_preview`
+- `image`
+- `image_count`
 
 ## Usage
 
-### 1) Load by `datasets`
-
 ```python
 from datasets import load_dataset
 
-# Replace with your dataset repo id
 repo_id = "g41/KnowCP"
 
-all_ds = load_dataset(repo_id, data_files={"train": "questions/questions_all.jsonl"})
-ir_ds = load_dataset(repo_id, data_files={"train": "questions/by_type/IR.jsonl"})
+gallery = load_dataset(repo_id, name="gallery")
+qa_ir = load_dataset(repo_id, name="qa_IR")
 
-print(all_ds["train"][0]["qid"])
-print(ir_ds["train"][0]["type"])
+print(gallery["train"][0]["image_id"])
+print(qa_ir["train"][0]["type"])
 ```
 
-### 2) Run evaluation with this project (`github_repo/run_eval_for_hf.py`)
+## Evaluation with This Project
 
-[GitHub link](https://github.com/41-edu/KnowCP)，This runner expects local files. If your dataset is already on Hugging Face, download it first and point `--workspace-root` to the parent folder that contains both `github_repo` and `hf_repo`.
+To run the full benchmark with local files:
 
 ```bash
-# Example local layout after download:
-# D:/mylib/benchmark/
-#   ├─ github_repo/
-#   └─ hf_repo/
-
 python github_repo/run_eval_for_hf.py \
 	--workspace-root D:/mylib/benchmark \
 	--items-dir D:/mylib/benchmark/hf_repo/questions/by_type \
@@ -123,21 +194,7 @@ python github_repo/run_eval_for_hf.py \
 	--api-key YOUR_API_KEY
 ```
 
-## Intended Uses
-
-- Benchmarking multimodal models on Chinese painting understanding
-- Evaluating grounding, OCR-style extraction, and text reasoning
-- Multi-turn visual QA experiments
-
-## Limitations
-
-- Prompts are optimized for benchmark consistency, not for general chat behavior.
-- Some tasks require strict JSON output formatting.
-- Evaluation metrics and threshold settings may differ by task type.
-
 ## Citation
 
-If you use this dataset, please cite the project repository and dataset page:
-
 - Code: https://github.com/41-edu/KnowCP
 - Dataset: https://huggingface.co/datasets/g41/KnowCP

viewer/qa_index.jsonl

@@ -0,0 +1,14 @@
+{"source_file": "CC.jsonl", "source_rows": 922, "preview_rows": 800, "preview_note": "Top 800 rows sampled for viewer"}
+{"source_file": "ER_choice.jsonl", "source_rows": 2875, "preview_rows": 800, "preview_note": "Top 800 rows sampled for viewer"}
+{"source_file": "ER_fillin.jsonl", "source_rows": 1279, "preview_rows": 800, "preview_note": "Top 800 rows sampled for viewer"}
+{"source_file": "IR.jsonl", "source_rows": 2351, "preview_rows": 800, "preview_note": "Top 800 rows sampled for viewer"}
+{"source_file": "ITT_MHQA_choice.jsonl", "source_rows": 5290, "preview_rows": 800, "preview_note": "Top 800 rows sampled for viewer"}
+{"source_file": "ITT_MHQA_fillin.jsonl", "source_rows": 5290, "preview_rows": 800, "preview_note": "Top 800 rows sampled for viewer"}
+{"source_file": "MITT_MHQA_choice.jsonl", "source_rows": 330, "preview_rows": 330, "preview_note": "Top 330 rows sampled for viewer"}
+{"source_file": "MITT_MHQA_fillin.jsonl", "source_rows": 330, "preview_rows": 330, "preview_note": "Top 330 rows sampled for viewer"}
+{"source_file": "PR.jsonl", "source_rows": 922, "preview_rows": 800, "preview_note": "Top 800 rows sampled for viewer"}
+{"source_file": "SR.jsonl", "source_rows": 1792, "preview_rows": 800, "preview_note": "Top 800 rows sampled for viewer"}
+{"source_file": "TR_choice.jsonl", "source_rows": 1312, "preview_rows": 800, "preview_note": "Top 800 rows sampled for viewer"}
+{"source_file": "TR_fillin.jsonl", "source_rows": 1312, "preview_rows": 800, "preview_note": "Top 800 rows sampled for viewer"}
+{"source_file": "TTI.jsonl", "source_rows": 1210, "preview_rows": 800, "preview_note": "Top 800 rows sampled for viewer"}
+{"source_file": "VA.jsonl", "source_rows": 922, "preview_rows": 800, "preview_note": "Top 800 rows sampled for viewer"}