List your company

Rule-discovery on Blicket baseline (test)

86Task SuccessQwen3-32B
29.8444.425973.58Apr 2, 2026

Evaluation results

MethodLinks
Qwen3-32BPrompt=TiO2026.048612.981.31
Qwen3-32B + 32B TiOPrompt=Base, Training=Distilled with Wason Task2026.047711.020.88
Qwen3-32BPrompt=Base2026.045712.180.54
Qwen3-8BPrompt=TiO2026.044010.730.57
Qwen3-8B + 32B TiOPrompt=Base, Training=Distilled with Wason Task2026.043811.840.51
Qwen3-8B + 32BPrompt=Base, Training=Distilled with Wason Task2026.043610.250.53
Qwen3-8B + 8B TiOPrompt=Base, Training=Distilled with Wason Task2026.043413.960.48
Qwen3-8BPrompt=Base2026.043210.260.44