Update dataset card for annotation-only release

README.md

@@ -10,39 +10,31 @@ size_categories:
 
 # MMCS 773K Data
 
-This repository contains the image archives and generated annotations used by MMCS (Multi-Modal Code-Switching).
+This repository contains the generated annotations used by MMCS (Multi-Modal Code-Switching). Source images are not redistributed here.
 
 ## Contents
 
-- `annotations/`: JSON annotation files with conversations format.
-- `images/`: ZIP archives grouped by source dataset.
+- `annotations/`: six JSON annotation files with conversations format. Each file is a JSON array, not JSONL.
 
-The dataset contains **773,779** samples. The ZIP entry paths exactly match the `image` fields in the annotations.
+The dataset contains **773,779** samples.
 
 | Source | Samples |
 | --- | ---: |
 | COCO 2014 | 75,033 |
 | Flickr30K | 29,140 |
 | GQA | 108,516 |
-| Objects365 part 1 | 252,988 |
-| Objects365 part 2 | 89,758 |
+| Objects365 | 342,746 |
 | Open Images | 127,237 |
 | SA-1B | 91,107 |
 | **Total** | **773,779** |
 
-## Download and extraction
+## Download
 
 ```bash
-hf download LockOnN/MMCS-Data --repo-type dataset --local-dir MMCS-Data
-
-mkdir -p data/images
-for archive in MMCS-Data/images/*.zip; do
-    unzip -q "$archive" -d data/images
-done
+hf download LockOnN/MMCS-Data \
+    --repo-type dataset \
+    --include "annotations/*" \
+    --local-dir MMCS-Data
 ```
 
-After extraction, configure the annotation paths under `annotations/` and use `data/images` as the image root.
-
-## Source data and licenses
-
-The images originate from COCO, Flickr30K, GQA, Objects365, Open Images, and SA-1B. Users are responsible for complying with the licenses and terms of the corresponding source datasets. The generated MMCS annotations do not replace or modify those source terms.
+See the [MM-CodeSwitch README](https://github.com/Changhao-Xiang/MM-CodeSwitch#data) for the complete data preparation and reproduction instructions.