List your company

ResearchFormerly WizWand

HS

Benchmarks

TaskDatasetSOTA resultTrendResults

Narrative Understanding

HS

87.8First-Token Accuracy

24

Commonsense Reasoning

HS

85.87Accuracy

23

Human Safety / Alignment

HS

1.84HS Average

18

Code Generation

HS (test)

67.1BLEU

10

Query representation clustering

HS (test)

0.072ARI

3