benchmark
WinoGrande
Metadata
| Source Table | benchmarks |
| Source ID | pIjI7HLxmb |
| Description | A large-scale commonsense reasoning benchmark with 44,000 Winograd-schema-style problems, using adversarial filtering to reduce annotation artifacts. |
| Wiki ID | winogrande |
| Children | — |
| Created | Mar 24, 2026, 11:23 PM |
| Updated | Mar 24, 2026, 11:24 PM |
| Synced | Mar 24, 2026, 11:24 PM |
Record Data
id | pIjI7HLxmb |
slug | winogrande |
name | WinoGrande |
category | reasoning |
description | A large-scale commonsense reasoning benchmark with 44,000 Winograd-schema-style problems, using adversarial filtering to reduce annotation artifacts. |
website | — |
scoringMethod | accuracy |
higherIsBetter | Yes |
introducedDate | 2019-07 |
maintainer | AI2 |
source | arxiv.org/abs/1907.10641 |
Debug info
Thing ID: pIjI7HLxmb
Source Table: benchmarks
Source ID: pIjI7HLxmb
Wiki ID: winogrande