Upload folder using huggingface_hub

README.md

@@ -1,7 +1,49 @@
 ---
 license: apache-2.0
 task_categories:
-- summarization
+  - summarization
 language:
-- en
----
+  - en
+---
+
+# 📚 Clip-CC Dataset
+
+**Clip-CC** is a high-quality dataset of 200 short educational videos (1 minute 30 seconds each) paired with human-written summaries. It’s designed for tasks like video summarization, video-language alignment, and multimodal learning.
+
+---
+
+## 🧾 Dataset Summary
+
+- 🎞️ 200 videos: Named `1.mp4` to `200.mp4`s
+- 📝 Each video has a concise human-written summary
+- 📂 File references stored in a JSONL file (`data.jsonl`) with paths to each video
+
+---
+
+## 💡 Use Cases
+
+- Video summarization and caption generation
+- Vision-Language alignment
+- Video QA and downstream educational AI tasks
+- Fine-tuning multimodal models (e.g., Flamingo, Video-BERT, LLaVA)
+
+---
+
+## 📦 Dataset Structure
+
+Each entry has:
+
+| Field     | Description                              |
+|-----------|------------------------------------------|
+| `id`      | Unique ID, e.g., `video_001`             |
+| `video`   | Relative path to video file              |
+| `summary` | Human-written summary of video content   |
+
+### 🔍 Example
+
+```json
+{
+  "id": "video_001",
+  "video": "videos/1.mp4",
+  "summary": "An introduction to gravitational force using animations."
+}