PALL-VLM: Dental Vision-Language Dataset for Instruction Tuning

32,884 records over 52,461 dental images formatted as image-text conversations for LLaVA-style instruction tuning. Curated by Harisundar R, this dataset serves as the training data for the PALL-VLM model. The dataset was last updated on June 12, 2026.

Use Cases

• Training dental vision-language models based on the 52,461 dental images and accompanying text conversations.
• Fine-tuning models for medical instruction following based on the LLaVA-style conversation format.
• Benchmarking multimodal AI performance in clinical visual question answering scenarios.

Strengths

• Contains 52,461 dental images, providing a substantial visual corpus.
• Includes 32,884 structured records formatted for instruction tuning.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count for individual files is unknown, which may limit suitability assessment.
• Data may reflect bias inherent to its source and curation process.

Provenance

Source

Harisundar R

Freshness

Last updated 2026-06-12 17:18:16; freshness should be verified.