Italian Medical Exam Questions for LLM Evaluation

Tommaso Mario Buonocore released novel datasets of Italian pre-university and post-university medical exam questions as part of a comparative evaluation of large language models. The datasets consist of five-choice questions covering clinical and preclinical fields, stored in an XLS file of 5.5 KB. The data was published on figshare in April 2026 under a CC-BY-4.0 license.

Use Cases

• Benchmarking LLM performance on specialized medical knowledge based on five-choice exam questions.
• Comparing multilingual versus Italian-specific model accuracy in a healthcare domain.
• Investigating the impact of model size versus domain specialization on task performance.
• Evaluating the pretrain-then-finetune paradigm using novel, unpublished test datasets.

Strengths

• Dataset is novel and unpublished, providing a unique benchmark for Italian medical LLM evaluation.
• Questions cover both clinical and preclinical fields from national pre-university and post-university exams.
• Released under a CC-BY-4.0 license, facilitating open research use.
• File size is 5.5 KB, indicating a focused and manageable dataset.

Limitations

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

Provenance

Source

Tommaso Mario Buonocore

Collection Method

Likely compiled from national pre-university and post-university medical exams.

Freshness

Last updated 2026-04-24 17:40:06; freshness should be verified.

Geography

Italy