Hierarchical Visual Question Answering for Emotion and Cognition

InsightVQA is a large-scale benchmark for hierarchical visual question answering that connects emotion understanding with cognitive reasoning. The dataset, created by ziyul707 and last updated in April 2026, is designed to evaluate model capabilities in interpreting emotional causes, grounding evidence, and performing reasoning.

Use Cases

• Train models to answer hierarchical questions linking visual scenes to emotional states and their cognitive causes.
• Benchmark model performance on tasks requiring evidence grounding for emotional interpretations within images.
• Evaluate reasoning capabilities on questions about emotional causes and consequences in visual contexts.

Strengths

• Designed as a large-scale benchmark for a structured evaluation paradigm.
• Last updated in April 2026, indicating recent maintenance.

Limitations

• Unknown total number of question-answer pairs and images.
• Potential geographic or cultural bias in emotional and cognitive labeling is unspecified.

Provenance

Source

huggingface

Collection Method

null

Time Range

null

Freshness

Last updated on 2026-04-06.

Geography

null