List your company

Mathematical Reasoning on GSM8K (test) (Accuracy Only)

59AccuracyNUDGING
27.835.94452.1Apr 22, 2026

Evaluation results

MethodLinks
NUDGINGPair=L→G, Intervention Rate=10.0%2026.0459
BlendInPair=L→G, Intervention Rate=10.0%2026.0459
NUDGINGPair=Q→G, Intervention Rate=23.5%2026.0450
BlendInPair=Q→G, Intervention Rate=23.5%2026.0450
NUDGINGPair=Q→L, Intervention Rate=22.4%2026.0429
BlendInPair=Q→L, Intervention Rate=22.4%2026.0429