Datasets:
Add task categories, relevant tags, paper link, and sample usage for Paper2Video dataset
#3
by
nielsr
HF Staff
- opened
This PR addresses several improvements for the Paper2Video dataset card:
- Metadata Enhancement:
- Adds
task_categories: ['any-to-any']to classify the dataset's domain. - Adds
tags: ['multimodal', 'video-generation']to better reflect the dataset's content and purpose. - Adds
language: ['en']as the dataset and documentation are in English.
- Adds
- Title Correction: Fixes the typo in the title from "Scientifuc Papers" to "Scientific Papers".
- Paper Link Update: Updates the paper link to directly point to the Hugging Face paper page (https://huggingface.co/papers/2510.05096).
- Sample Usage Section: Incorporates a "Sample Usage" section, extracted from the official GitHub README, which includes:
- Environment setup commands.
- Instructions for configuring LLM API credentials.
- The main Python command (
pipeline.py) to run the video generation process, providing users with a clear starting point.
These changes will improve the discoverability, accuracy, and usability of the Paper2Video dataset on the Hugging Face Hub.