List your company

Instruction Following on IF

39.2IF ScoreChat-Math-IF
24.84828.57432.336.026Feb 2, 2026

Evaluation results

MethodLinks
Chat-Math-IFBackbone=Qwen-2.5-7B, Training Strategy=SEQUENTIAL Training2026.0239.240.3
Modular Gradient SurgeryBackbone=Qwen-2.5-7B, Training Strategy=MIXED Training2026.0233.745
Normal MixingBackbone=Qwen-2.5-7B, Training Strategy=MIXED Training2026.0233.244.6
Global SurgeryBackbone=Llama-3.1-8B, Training Strategy=MIXED Training2026.0231.130.3
Modular Gradient SurgeryBackbone=Llama-3.1-8B, Training Strategy=MIXED Training2026.023032.6
Global SurgeryBackbone=Qwen-2.5-7B, Training Strategy=MIXED Training2026.0229.943
Math-Chat-IFBackbone=Qwen-2.5-7B, Training Strategy=SEQUENTIAL Training2026.0226.220.2
Normal MixingBackbone=Llama-3.1-8B, Training Strategy=MIXED Training2026.0225.427.3