Table 5

Performance results of models (Qwen2.5–72B, Qwen2.5-72B-finetuning, Qwen2.5-max and Qwen3-235B)

TasksModels
Qwen2.5–72B baseQwen2.5-72B-finetuningQwen2.5-max baseQwen3-235B-A22 B base
PRF1PRF1PRF1PRF1
MNR0.6910.6810.6860.7410.7140.7280.6490.7180.6820.6460.6560.651
DIA0.5390.7240.6180.7960.6030.6860.5630.7760.6520.6670.7590.710
ETI0.4290.6000.5000.3910.9000.5460.3480.8000.4850.3500.7000.467
GMI0.4710.5000.4850.4380.4380.4380.5330.5000.5160.3000.3750.333
PROG0.3330.2940.3130.2220.2350.2290.2000.2940.2380.2220.3530.273
TREAT0.8620.7270.7890.8640.8140.8380.8180.7560.7860.8060.6740.734
ENR0.3820.4620.4180.4370.4530.4440.4050.4930.4450.4940.5200.507
FEEL0.4600.5120.4850.4300.4220.4260.4740.5200.4960.5840.5940.589
VIEW0.3000.3980.3420.4440.4900.4660.3360.4600.3880.3890.4290.408
MNCX0.6440.6480.6460.7100.6720.6900.7090.7110.7100.6580.7260.690
BACK0.5930.5090.5490.7290.5560.6310.7020.6560.6780.6320.6550.643
CON0.4310.5000.4630.5000.4670.4830.4190.6050.4950.4670.6360.539
ELA0.7450.8350.7870.7420.8680.8000.8240.7950.8100.7500.8270.787
ENCX (CAUSE)0.7050.7330.7180.8380.7950.8160.7710.7710.7710.7950.8090.802
Overall0.5950.6240.6090.6340.6240.6290.6230.6710.6460.6330.6690.651
Source(s): Authors’ own work

or Create an Account

Close Modal
Close Modal