SoccerNet VQA: Multimodal Question Answering for 14 Soccer Tasks

The dataset supports the 2026 Soccernet Challenge for multimodal (text, image, video) multiple-choice question answering. It covers 14 distinct soccer understanding tasks, including assessing player and team background knowledge, determining camera status, classifying actions, and recognizing fouls. The dataset was created by SoccerNet and last updated in October 2025.

Use Cases

• Train multimodal models on multiple-choice questions covering 14 distinct soccer understanding tasks.
• Benchmark action classification and foul recognition models using video and text data.
• Assess model background knowledge of players and teams through multimodal question answering.
• Develop models to determine camera status from video and image inputs.

Strengths

• Covers 14 distinct soccer understanding tasks for multimodal analysis.
• Created by SoccerNet, a known organization for soccer analytics data.
• Last updated in October 2025, indicating recent maintenance.

Limitations

• Specific data volume, column structure, and sample data are unavailable for assessment.
• The multimodal nature (text, image, video) may require specialized tools and significant computational resources for processing.
• Potential for class or task imbalance across the 14 distinct understanding tasks is unknown.

Provenance

Source

SoccerNet

Freshness

2025-10-18