# LearningPaper24 Dataset

This dataset contains video recordings and metadata from ICLR and NIPS 2024 conference talks. It includes both poster and oral presentations, along with their associated metadata such as titles, abstracts, and keywords.

## Dataset Structure

```
learningpaper24/
├── README.md
├── metadata/
│   └── catalog.json
└── video/
    ├── {openreview_id}_{slideslive_id}.mp4
    └── ...
```

## Data Format

### Catalog (metadata/catalog.json)
The catalog contains metadata for each talk in JSON format with the following fields:
- `video_file`: Filename of the video recording in the format `{openreview_id}_{slideslive_id}.mp4`
- `openreview_id`: Unique identifier from OpenReview
- `slideslive_id`: Video identifier from SlidesLive
- `venue`: Conference venue (e.g., "iclr2024")
- `title`: Paper title
- `status`: Presentation type (e.g., "Poster", "Oral")
- `keywords`: Research keywords
- `tldr`: Short summary
- `abstract`: Full paper abstract
- `primary_area`: Main research area
- `site`: Link to the conference page

### Videos
Videos are stored in the `video` directory with filenames following the format: `{openreview_id}_{slideslive_id}.mp4`

## Purpose

This dataset can be used for:
- Video understanding and summarization
- Natural language processing tasks
- Video-text alignment studies

<!-- ## Dataset Statistics

[To be added: Include statistics about number of videos, duration, distribution of presentation types, etc.]

!-->

## License

This dataset is released under the [Creative Commons Attribution-NonCommercial 4.0 International License](https://creativecommons.org/licenses/by-nc/4.0/).

<!-- ## Citation

If you use this dataset, please cite:
```
@misc{learningpaper24,
  author = {[Your Name]},
  title = {LearningPaper24: A Dataset of ICLR and NIPS 2024 Conference Talks},
  year = {2024},
  publisher = {Hugging Face},
  journal = {Hugging Face Hub},
  howpublished = {\url{https://huggingface.co/datasets/[username]/learningpaper24}}
}
``` -->