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

Dialogue Evaluation Alignment on CRSArena-Eval RD (test)

0.549Pearson r (Turn-level Rel)

FACE

0.301480.365740.430.49426Jun 11, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
0.5490.550.4430.4370.650.6350.570.4530.4840.5340.4470.430.7120.6680.5510.53
2026.06
0.490.4710.3020.2890.490.4440.3510.3640.4880.4820.3320.3250.5770.5770.4330.422
2026.06
0.4680.4620.2790.290.6050.5740.4820.3920.3390.4230.2350.2550.6170.5550.4320.422
2026.06
0.4640.4550.2480.260.5220.4820.4050.3630.1010.1010.2170.2030.5640.5220.360.341
2026.06
0.4530.4460.1750.1770.4810.4570.4250.40.1740.1740.1880.1740.4980.4720.3420.329
2026.06
0.3110.2880.1820.2420.2460.225------0.3950.387--