Turkish Image Captioning Dataset Translated from BLIP3o Pretraining Data

A subset of the BLIP3o-Pretrain-Long-Caption and BLIP3o-Pretrain-Short-Caption datasets translated into Turkish. The dataset is intended for training or fine-tuning image-to-text models. It was created by the author 'ituperceptron' and was last updated on January 15, 2026.

Use Cases

• Fine-tuning image-to-text models based on Turkish-language captions.
• Training Turkish-language image captioning models based on translated pretraining data.
• Evaluating the performance of multimodal models on Turkish-language tasks.
• Studying the effects of machine translation on caption quality for model training.

Strengths

• Derived from established pretraining datasets (BLIP3o-Pretrain-Long-Caption and BLIP3o-Pretrain-Short-Caption).
• Provides Turkish-language captions, a resource for a less common language in this domain.
• Likely contains paired image and text data suitable for multimodal learning.

Limitations

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

Provenance

Source

Translated subset of BLIP3o pretraining datasets.

Collection Method

Machine translation of captions from original datasets.

Freshness

Last updated 2026-01-15 05:12:49; freshness should be verified.