CommonsenseQA Benchmark

Explore score history ↗

CommonsenseQA is an AI benchmark in the Benchmark Radar catalog.

Released: 2018-11-05
Openness: unknown
Importer: Benchmark Saturation

Source provenance

Original evidence

CommonsenseQA results and reported scores

Reported benchmark scores · Benchmark Saturation

Reported scores: 5

Model	Organization	Reported value	Reported	Evidence
DeBERTaV3-large+KEAR	—	91.2	—	evidence
ESIM+ELMo	—	34.1	—	evidence
GPT	—	45.5	—	evidence
Mistral NeMo Instruct	—	70.4	—	evidence
Palm 2	—	90.4	—	evidence

Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.

Which sources cite CommonsenseQA?

CommonsenseQA