Rare Species Images and Text for Zero-Shot Classification

A multimodal dataset of rare species images and text descriptions, generated alongside the TreeOfLife-10M dataset. Data was pulled from the Encyclopedia of Life (EOL) by the Imageomics organization. The dataset was last updated on Hugging Face in January 2026.

Use Cases

• Zero-shot classification of biological species based on image and text data.
• Training or fine-tuning vision-language models for taxonomic tasks.
• Benchmarking computer vision models on rare species recognition.
• Supporting research in biodiversity and computational ecology.

Strengths

• Data is sourced from the authoritative Encyclopedia of Life (EOL).
• Dataset is associated with the BioCLIP vision foundation model research paper.
• Last update timestamp is explicitly provided (2026-01 09 16:26:20).

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 and file size are unknown, which may limit suitability assessment.

Provenance

Source

Encyclopedia of Life (EOL)

Collection Method

Data (images and text) were pulled from EOL.

Time Range

null

Freshness

Last updated 2026-01-09 16:26:20.

Geography

null