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

Question Answering on ConflictQA

53.8Accuracy

CR-DPO

-1.73612.68227.141.518May 11, 2026
Updated 2mo ago

Evaluation Results

MethodLinks
2026.05
53.894.981.688.2
2026.05
53.5978591
2026.05
49.699.39094.6
2026.05
47.894.53665.3
2026.05
47.898.56179.8
2026.05
47.389.536.162.8
2026.05
4792.734.763.7
2026.05
46.997.982.990.4
2026.05
46.81007.553.7
2026.05
44.5100753.5
2026.05
44.3962761.5
2026.05
449875.486.7
2026.05
43.492.919.356.1
2026.05
42.990.112.251.2
2026.05
42.1100753.5
2026.05
421005.452.7
2026.05
41.9841348.5
2026.05
41.696.740.468.5
2026.05
41.492.437.865.1
2026.05
41.388.120.354.2
2026.05
40.973.952.463.1
2026.05
39.879.285.682.4
2026.05
39.590.93.547.2
2026.05
38.692.231.661.9
2026.05
38.272.16367.6
2026.05
37.274.16569.5
2026.05
36.782.96.244.6
2026.05
34.870.73050.3
2026.05
31.937.672.154.9
2026.05
31.878.41446.2
2026.05
31.736.679.658.1
2026.05
3155.95053
2026.05
30.664.957.961.4
2026.05
29.551.968.160
2026.05
27.654.668.561.6
2026.05
22.7310051.5
2026.05
15.94.470.937.7
2026.05
14.61.310050.6
2026.05
13.90.810050.4
2026.05
12.313.870.242
2026.05
10.52.628.315.5
2026.05
10.46.446.226.3
2026.05
9.11.110050.6
2026.05
0.400.90.5