Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Prompt Optimization Evaluation on HelpSteer 1

51.83Helpfulness

MA-SAPO

22.460430.085237.7145.3348Oct 18, 2025
Updated 2mo ago

Evaluation Results

MethodLinks
2025.10
51.8362.686.1450.1373.6364.86
2025.10
41.5948.7679.6332.9351.8850.96
2025.10
41.148.6883.2647.284.3360.91
2025.10
39.8846.7980.2434.2750.0850.25
2025.10
39.0145.9177.4535.1258.0151.1
2025.10
37.7447.6479.5432.152.4849.9
2025.10
37.7450.4980.6740.8467.0855.31
2025.10
37.5144.0278.7131.1647.7947.84
2025.10
32.2338.7675.9529.3541.9243.64
2025.10
32.1638.6675.8329.5142.1543.66
2025.10
29.335.9474.5230.0743.7742.72
2025.10
28.8735.1672.8731.6445.1742.74
2025.10
25.4932.4770.5427.0240.6239.27
2025.10
23.5930.7769.0626.2739.6737.87