List your company

Question Answering on TRQA (test)

51.16Initial ScoreQWEN3-MAX
22.747230.123637.544.8764May 11, 2026

Evaluation results

MethodLinks
QWEN3-MAXModel Scale=Max2026.0551.16-2.334.656.9848.84
DPOModel Scale=9B2026.0533.721.162.911.7434.88
BASE MODELModel Scale=9B2026.0531.98-0.581.742.3331.4
GRPOModel Scale=9B2026.0531.981.743.491.7433.72
ReCritModel Scale=9B2026.0531.989.314.535.2341.28
CRITIQUEGRPOModel Scale=9B2026.0531.42.915.232.3334.3
SFTModel Scale=9B2026.0530.811.162.911.7431.98
GRPOModel Scale=4B2026.0527.330.581.160.5827.91
DPOModel Scale=4B2026.0526.16-0.580.581.1625.58
BASE MODELModel Scale=4B2026.0525-1.1601.1623.84
SFTModel Scale=4B2026.0524.42-1.741.162.9122.67
ReCritModel Scale=4B2026.0524.4211.0518.026.9835.47
CRITIQUEGRPOModel Scale=4B2026.0523.842.915.232.3326.74