DROP

DROP is a benchmark for general reasoning and comprehension. The official description says: "DROP - discrete reasoning over paragraphs". Below we describe in simple words what benchmarks of this type generally examine.

DROP measures general reasoning and comprehension. In detail, it checks:

• drawing the right conclusion from a passage or scenario
• commonsense judgement about everyday situations and words
• counting, comparing and reasoning over several facts at once
• avoiding shortcuts that memorise answers instead of reasoning

The score is accuracy: the percentage of tasks answered or solved correctly. 100% means every task was solved perfectly; higher is better.

Official methodology note: DROP - discrete reasoning over paragraphs. 1-shot.

What is the DROP benchmark?

DROP is a benchmark for general reasoning and comprehension. The official description says: 'DROP - discrete reasoning over paragraphs'. Below we describe in simple words what benchmarks of this type generally examine.

What does DROP test?

DROP mainly measures general reasoning and comprehension. Concretely, it checks: drawing the right conclusion from a passage or scenario; commonsense judgement about everyday situations and words; counting, comparing and reasoning over several facts at once.

How is the DROP score calculated?

The score is accuracy: the percentage of tasks answered or solved correctly. 100% means every task was solved perfectly; higher is better. The declared maximum score of this benchmark is 100.0.

Where do I get DROP data and scores?

The official source is https://arxiv.org/abs/1903.00161. The official benchmark page/reference is linked on this page. Every score on Open LLM Radar is also available as open data: download the scores JSON (http://open-llm-radar-api:8000/api/v1/benchmarks/drop/scores) or pull it directly from the Open LLM Radar API with any HTTP client.