List your company
ResearchFormerly WizWand

Distractor Generation on CLOTH

73.25Hardest AccuracyDCDG + ASDE, DDDE
4.048422.014239.9857.9458Nov 3, 2025
Updated 5mo ago

Evaluation results

MethodLinks
DCDG + ASDE, DDDEType=Hard2025.1173.2519.247.31
DCDG + ASDEType=Hard2025.1172.1419.747.62
DCDGType=Hard2025.1170.35209.15
DCDG + DDDEType=Hard2025.1167.9723.398.73
GPT-4o (0-shot)Type=Hard2025.1156.7728.0815.15
GPT-4o (5-shot)Type=Hard2025.1153.8130.5615.83
GPT-4o (0-shot)Type=Easy2025.1129.636.3633.54
GPT-4o (5-shot)Type=Original2025.1124.3538.3837.78
DCDG + DDDEType=Original2025.1122.6943.9833.73
GPT-4o (5-shot)Type=Easy2025.1121.8431.0646.39
DCDGType=Original2025.1120.447.9432.36
DCDG + ASDE, DDDEType=Original2025.1120.0451.828.46
DCDG + ASDEType=Original2025.1119.6449.731.66
GPT-4o (0-shot)Type=Original2025.1113.6435.5651.31
DCDG + DDDEType=Easy2025.119.3432.6357.53
DCDGType=Easy2025.119.2532.0658.49
DCDG + ASDEType=Easy2025.118.2230.5660.72
DCDG + ASDE, DDDEType=Easy2025.116.7128.9664.23