Human-Subject Experiments on AI Image Caption Evaluation with Informational Cues

Jiang, Yanru provides experimental materials and anonymized participant data supporting the paper 'Informational Cues Mitigate Heuristic Bias in Evaluating Context-Sensitive AI Image Captions'. The data documents two human-subject experiments examining how informational context shapes audience evaluations of AI-generated image captions. The dataset was last updated on July 14, 2026.

Use Cases

• Study the impact of informational context on human evaluation of AI outputs based on the described experiments.
• Analyze heuristic bias mitigation in multimodal AI evaluation based on the experimental design.
• Replicate or extend the findings on context-sensitive caption evaluation based on the provided materials.

Strengths

• Includes anonymized participant data from two human-subject experiments.
• Contains experimental design materials and computational analyses supporting a specific research paper.
• Last updated on July 14, 2026, indicating recent maintenance.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment for large-scale analyses.

Provenance

Source

Harvard Dataverse, author JIANG, YANRU

Collection Method

Data from two human-subject experiments, likely collected in a controlled setting.

Freshness

Last updated 2026-07-14 21:33:00