MBE2.0: E-Commerce Product Images, Titles, and Annotations

A Chinese multimodal benchmark for e-commerce product understanding, released following a legal and privacy review aligned with China's PIPL. The dataset includes original images, product titles, and category/attribute annotations, with all personally identifiable information removed. It was created by author ZHNie and last updated on March 23, —.

Use Cases

• Train multimodal representation learning models based on the paired image and text data.
• Benchmark product categorization systems based on the provided category annotations.
• Develop attribute prediction models for e-commerce products based on the attribute annotations.
• Conduct research on modality-balanced learning for e-commerce tasks.

Strengths

• Includes original images, titles, and annotations, providing a complete multimodal source.
• Personally identifiable information has been rigorously removed, addressing privacy concerns.
• Released following a formal legal, privacy, and compliance review process.

Limitations

• Row count, file formats, and column-level documentation are unknown, which may limit suitability assessment.
• The training set and model weights are noted as undergoing final security review and are not yet fully released.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

HuggingFace repository by author ZHNie, associated with the paper 'MOON2.0: Dynamic Modality-balanced Multimodal Representation Learning for E-commerce Product Understanding'.

Collection Method

Likely gathered for research purposes; specific collection method is not detailed.

Time Range

null

Freshness

Last updated 2026-03-23 03:40:59; freshness should be verified.

Geography

Data processing and release aligned with China's PIPL law, suggesting a Chinese context.