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

Dialogue Evaluation Alignment on CRSArena-Eval KG (test)

0.543Pearson r (Turn-level Rel.)

FACE

0.410920.445210.47950.51379Jun 11, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
0.5430.5270.4710.4530.7190.6770.5930.4840.5180.5430.4490.4040.7660.6790.580.538
2026.06
0.5330.5050.3340.3160.5350.4750.430.4220.4850.4630.4240.4030.6560.6420.4850.461
2026.06
0.4920.4860.3080.3240.6640.6110.4260.4110.240.4190.2970.3220.6720.5570.4430.447
2026.06
0.4520.4520.2380.2310.5990.5460.5380.4810.1370.1370.4250.3780.6550.5570.4350.397
2026.06
0.4190.4080.2030.190.5620.520.4750.4340.1140.1140.3090.2790.5990.5210.3830.352
2026.06
0.4160.4280.2620.4010.5630.541------0.6180.659--