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

Multi-way preference ranking on EDITREWARD-BENCH

66.2Preference Score (K=2)

REWARDHARNESS (Gemini-2.0-Flash)

24.194435.099746.00556.9103Sep 30, 2025Nov 5, 2025Dec 12, 2025Jan 18, 2026Feb 24, 2026Apr 2, 2026May 9, 2026
Updated 22d ago

Evaluation Results

MethodLinks
2026.05
66.245.313.5-
2025.09
58.6139.8612.1638.02
2026.05
58.639.912.2-
2026.05
57.930.77.4-
2026.05
57.946.710.8-
2025.09
57.5338.5112.8437.81
2026.05
57.538.512.8-
2026.05
573610.8-
2025.09
56.993610.8136.78
2026.05
56.542.711.5-
2025.09
56.4542.6711.4938.42
2026.05
52.724.73.4-
2025.09
52.6924.673.3829.75
2025.09
52.4333.3313.5133.47
2026.05
52.433.313.5-
2025.09
51.0720.272.7126.86
2025.09
50.5425.274.0528.72
2026.05
50.525.34.1-
2026.05
49.530.49.5-
2025.09
49.4630.419.4631.19
2026.05
45.727.37.3-
2025.09
45.6927.337.3128.31
2025.09
25.8111.331.3513.84