Human Style Preferences Images: Over 1.2 Million Annotations for Text-to-Image Models

One of the largest human preference datasets for text-to-image models, containing over 1,200,000 human preference annotations. It was collected by Rapidata using their Python API over approximately four days. The dataset was last updated on January 10, 2025.

Use Cases

• Training reward models for text-to-image generation based on human preference annotations.
• Benchmarking the aesthetic quality of different image generation models.
• Analyzing human stylistic preferences across a large set of generated images.
• Fine-tuning image generation models to better match human taste.

Strengths

• Contains over 1,200,000 human preference annotations, indicating a large scale.
• Data collection was completed in approximately four days using a dedicated API, suggesting a rapid annotation process.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Rapidata

Collection Method

Collected using the Rapidata Python API for large-scale data annotation.

Freshness

Last updated 2025-01-10 21:59:31; freshness should be verified.