List your company

LLM-judge evaluation on AXBENCH

92.5Concept ScoreSPLIT
19.738.657.576.4Feb 2, 2026

Evaluation results

MethodLinks
SPLITModel=Gemma-2-9B-IT, Form=Local Weight2026.0292.581.125
SPLITModel=Gemma-2-9B-IT, Form=Vector2026.0292.582.375
SPLITModel=Qwen-2.5-7B-IT, Form=Vector2026.0290.62582.5
REPSModel=Gemma-2-9B-IT, Form=Local Weight2026.0288.7581.81
SPLITModel=Gemma-2-9B-IT, Form=LoRA2026.0288.7582.06
SFTModel=Gemma-2-9B-IT, Form=LoRA2026.0288.12575.94
REPSModel=Gemma-2-9B-IT, Form=LoRA2026.0286.87582.625
SPLITModel=Qwen-2.5-7B-IT, Form=LoRA2026.0286.87581.81
REPSModel=Qwen-2.5-7B-IT, Form=Vector2026.0286.87582.06
SFTModel=Gemma-2-9B-IT, Form=Vector2026.028572.435
REPSModel=Gemma-2-9B-IT, Form=Vector2026.028577.75
SPLITModel=Qwen-2.5-7B-IT, Form=Local Weight2026.028571.625
REPSModel=Qwen-2.5-7B-IT, Form=Local Weight2026.0284.37570.815
SFTModel=Gemma-2-9B-IT, Form=Local Weight2026.0283.12572.69
SFTModel=Qwen-2.5-7B-IT, Form=Vector2026.0278.7579
SFTModel=Qwen-2.5-7B-IT, Form=Local Weight2026.0276.87556.435
SFTModel=Qwen-2.5-7B-IT, Form=LoRA2026.0274.37565.875
REPSModel=Qwen-2.5-7B-IT, Form=LoRA2026.0274.37570.065
DiffMeanModel=Gemma-2-9B-IT, Form=Vector2026.0258.12552.75
DiffMeanModel=Qwen-2.5-7B-IT, Form=Vector2026.0247.549.75
VanillaModel=Gemma-2-9B-IT, Form=Vanilla2026.0223.7524.75
VanillaModel=Qwen-2.5-7B-IT, Form=Vanilla2026.0222.523.565