README upt

README.md

@@ -2,23 +2,22 @@
 
 MoSu (Most Replayed Multimodal Video Summarization) is the first large-scale multimodal video summarization dataset. It provides synchronized visual, audio, and text features for 52,678 in-the-wild videos. The ground-truth annotations are based on YouTube's "Most Replayed" statistics, offering highly reliable per-frame importance scores derived from collective viewer engagement.
 
-This dataset is officially introduced in the ICLR 2026 paper, "TripleSumm: Adaptive Triple-Modality Fusion for Video Summarization".
 * **Paper:** [TripleSumm: Adaptive Triple-Modality Fusion for Video Summarization](https://arxiv.org/abs/2603.01169)
 * **GitHub Repository:** [smkim37/TripleSumm](https://github.com/smkim37/TripleSumm)
 
 ## Dataset Structure
 
-The dataset consists of 6 core files categorized into Metadata, Splits, Ground Truth, and Features.
+The dataset consists of 6 core files providing metadata, multimodal features, ground truth annotations, and evaluation splits.
 
 ### 1. Metadata (`mosu_metadata.csv`)
 
 Contains the foundational information for all 52,678 videos.
 
-* **video_id**: The unique identifier for the video. This serves as the universal key to access data in all **.h5** files and the split JSON.
-* **youtube_id**: The original YouTube video ID. The video can be accessed via [https://www.youtube.com/watch?v=](https://www.youtube.com/watch?v=){youtube_id}.
+* **video_id**: The unique identifier for the video. This serves as the universal key to access data in all .h5 files and the split JSON.
+* **youtube_id**: The original YouTube video ID. The video can be accessed via `https://www.youtube.com/watch?v={youtube_id}`.
 * **duration**: The length of the video in seconds.
 * **views**: The total view count of the video.
-* **labels**: Original multi-label annotations provided by the YouTube-8M dataset.
+* **labels**: Original multi-label annotations provided by the [YouTube-8M dataset](https://research.google.com/youtube8m/).
 * **cluster_id**: One of 10 semantic clusters (0-9). These clusters were generated based on metadata to group videos by topic (e.g., Video Games, Sports) and ensure a balanced distribution across dataset splits.
 
 ### 2. Multimodal Features (`.h5 files`)
@@ -34,10 +33,8 @@ All features have a shape of (N, D), where N corresponds to the video's duration
 
 ### 3. Ground Truth (`mosu_gt.h5`)
 
-An HDF5 file containing the summarization labels.
+An HDF5 file containing the summarization labels for all 52,678 videos. Each `video_id` (e.g., '005O') maps to an HDF5 Group containing four specific keys:
 
-* **Total Keys:** 52,678
-* **Data Structure:** Each **video_id** (e.g., '005O') maps to an HDF5 Group containing four specific keys:
 * **change_points**: Temporal boundaries for video shots.
 * **cluster_id**: The semantic cluster ID of the video.
 * **gt_score**: Frame-level ground-truth importance scores.
@@ -45,10 +42,11 @@ An HDF5 file containing the summarization labels.
 
 ### 4. Dataset Splits (`mosu_split.json`)
 
-Contains lists of **video_id**s designated for the train, validation, and test sets. The split ratio strictly maintains the proportional representation of each **cluster_id** for a balanced evaluation.
-* **train_keys**: 42,152 videos
-* **Val_keys**: 5,263 videos
-* **test_keys**: 5,263 videos
+Contains standardized splits for training, validation, and testing. The split ratio strictly maintains the proportional representation of each `cluster_id` for a balanced evaluation.
+
+* **train_keys**: List of video IDs for 42,152 training videos.
+* **val_keys**: List of video IDs for 5,263 validation videos.
+* **test_keys**: List of video IDs for 5,263 testing videos.
 
 ## Citation