SuperGLUE

Measures: Eight language-understanding tasks (question answering, inference, coreference, word sense) built to be harder than GLUE; scored as their average.

Maintainer: Wang et al. (NYU, Facebook AI, University of Washington, DeepMind)
Introduced: 2019
Status: Unmaintained
Saturation: Near-ceiling
Typical provenance: Self-reported
In scoring set: No

Paper

Leaderboard

Show all results

Independent and third-party results only.

Leaderboard
Model	Average score	Provenance	Source	Evaluated
text-davinci-001 (InstructGPT) OpenAI	71.8%	Third party	Language Models are Few-Shot Learners via Epoch AI	—

Epoch AI, 'Capabilities & benchmarking', epoch.ai/benchmarks, CC BY 4.0.

Sources

Epoch AI