Add README and .gitattributes

.gitattributes

@@ -57,3 +57,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 # Video files - compressed
 *.mp4 filter=lfs diff=lfs merge=lfs -text
 *.webm filter=lfs diff=lfs merge=lfs -text
+*.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,41 @@
+# RefCOCO JSON Files
+
+This repository contains both original and filtered RefCOCO, RefCOCO+, and RefCOCOg JSON files.
+
+## Contents
+
+### Original Files
+- refcoco_train.json
+- refcoco_val.json
+- refcoco_testA.json
+- refcoco_testB.json
+- refcocop_train.json
+- refcocop_val.json
+- refcocop_testA.json
+- refcocop_testB.json
+- refcocog_train.json
+- refcocog_val.json
+- refcocog_test.json
+- flickr_single_bbox_train.json
+- flickr_single_bbox_val.json
+- flickr_single_bbox_test.json
+- refgta_subsample.json
+
+### Filtered Files (Long Captions)
+- filtered/refcoco_train_long_captions.json
+- filtered/refcocop_train_long_captions.json
+- filtered/refcocog_train_long_captions.json
+
+## Usage
+
+The filtered JSON files contain a subset of the original data with longer, more descriptive captions. These are particularly useful for training models that benefit from richer linguistic descriptions.
+
+### Download and Setup
+
+You can use the `prepare_data.sh` script from the VLM-R1 repository to download and set up these files:
+
+```bash
+bash prepare_data.sh
+```
+
+This will download both the original and filtered JSON files and set up the appropriate YAML configuration files.