List your company

Text Revision on combined real-LLM corpus

10.74Mean Score (Comparator)Full-Restart
-0.42962.47025.378.2698Apr 25, 2026

Evaluation results

MethodLinks
Full-RestartMetric Category=waste2026.0410.74—10.74—102.0510.89
OracleMetric Category=quality2026.043.793.79——0.750.770.95
Full-RestartMetric Category=quality2026.043.163.16——0.110.120.27
NaiveMetric Category=quality2026.042.722.72——0.330.30.13
IgnoreMetric Category=quality2026.041.661.66——1.381.321.19
OracleMetric Category=waste2026.040—0—0.742.140.69
NaiveMetric Category=waste2026.040—0—0.742.380.69
AbsorberMetric Category=quality2026.04—3.04—3.04———
AbsorberMetric Category=waste2026.04——0.740.74———