| { |
| "@context": { |
| "@vocab": "https://mlcommons.org/croissant/", |
| "schema": "https://schema.org/", |
| "name": "schema:name", |
| "description": "schema:description", |
| "identifier": "schema:identifier", |
| "license": "schema:license", |
| "creator": "schema:creator", |
| "affiliation": "schema:affiliation", |
| "recordSets": "recordSets", |
| "fields": "fields", |
| "dataType": "dataType" |
| }, |
| "@type": "schema:Dataset", |
| "name": "worldmodelbench", |
| "description": "A benchmark for evaluating world modeling ability in video generation models.", |
| "identifier": "https://huggingface.co/datasets/Efficient-Large-Model/worldmodelbench", |
| "license": "CC BY 4.0", |
| "creators": [ |
| { |
| "name": "World Model Bench Team", |
| "affiliation": "Nvidia, MIT, UC Berkeley, UC San Diego" |
| } |
| ], |
| "recordSets": [ |
| { |
| "name": "benchmark_instances", |
| "description": "Each record includes a text instruction, an initial video frame, and associated metadata for evaluation.", |
| "fields": [ |
| { |
| "name": "text_instruction", |
| "description": "The instruction provided for the video generation task.", |
| "dataType": "https://schema.org/Text" |
| }, |
| { |
| "name": "text_first_frame", |
| "description": "Description of the initial video frame.", |
| "dataType": "https://schema.org/Text" |
| }, |
| { |
| "name": "first_frame", |
| "description": "Filename of the initial video frame image.", |
| "dataType": "https://schema.org/Text" |
| }, |
| { |
| "name": "domain", |
| "description": "The application domain of the prompt (e.g., Robotics, Driving).", |
| "dataType": "https://schema.org/Text" |
| }, |
| { |
| "name": "subdomain", |
| "description": "The specific subdomain within the application domain.", |
| "dataType": "https://schema.org/Text" |
| } |
| ] |
| } |
| ] |
| } |
|
|
|
|
