DIM-T2I: Text-to-Image and Image Editing Dataset for AI Models

DIM-Edit dataset accompanies the DIM-4.6B-T2I and DIM-4.6B-Edit models released in October 2025. The dataset supports research on rebalancing designer and painter roles in unified multimodal models for image editing. It was created by Ziyun Zeng, David Junhao Zhang, Wei Li, and Mike Zheng Shou, with the associated paper accepted to ICLR 2026.

Use Cases

• Training text-to-image generation models based on the DIM-4.6B-T2I model mentioned in the description
• Benchmarking image editing capabilities based on the DIM-4.6B-Edit model mentioned in the description
• Researching role rebalancing in multimodal architectures based on the paper's focus mentioned in the description

Strengths

• Dataset is associated with a peer-reviewed paper accepted to ICLR 2026
• Dataset is linked to specific model releases (DIM-4.6B-T2I and DIM-4.6B-Edit) from October 2025

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

Author is listed as 'stdKonjac' on Hugging Face.

Freshness

Last updated 2026-05-11 19:46:19; freshness should be verified