List your company

Instruction Following on ComplexBench (CSR, TCR, Conf. Res., Avg-T)

78.6CSRCRGC
66.22469.43772.6575.863Jun 2, 2026

Evaluation results

MethodLinks
CRGCBackbone=GPT-4o2026.0678.679.873.91.28
DeCRIMBackbone=GPT-4o2026.0675.176.854.24.2
SPBackbone=GPT-4o2026.0666.768.128.51