NSD-VQA: Visual Question Answering Benchmark from Human fMRI Responses

NSD-VQA is a large-scale visual question answering benchmark for studying the decoding of visual and semantic information from human fMRI responses to natural images. It is built from the Natural Scenes Dataset (NSD) and provides automatically generated question-answer annotations grounded in NSD images. The dataset was created by mcosarinsky and was last updated on 2026-05-24.

Use Cases

• Benchmarking brain activity decoding models based on fMRI responses to images.
• Training models for visual question answering based on automatically generated annotations.
• Studying the relationship between visual stimuli and semantic understanding based on fMRI data.

Strengths

• Dataset is described as a 'large-scale' benchmark, suggesting substantial size.
• Annotations are automatically generated and grounded in the established Natural Scenes Dataset images.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and dataset size are unknown, which may limit suitability assessment.

Provenance

Source

Built from the Natural Scenes Dataset (NSD).

Collection Method

Provides automatically generated question-answer annotations grounded in NSD images.

Freshness

Last updated 2026-05-24 08:13:04; freshness should be verified.