List your company

Mathematical Reasoning on AIME24 search

85.8AccuracyAutoTTS (β=1.0)
16.53634.51852.570.482May 8, 2026

Evaluation results

MethodLinks
AutoTTS (β=1.0)Base Model=Qwen3-8B, Type=Discovered2026.0585.8467,400
AutoTTS (β=0.5)Base Model=Qwen3-8B, Type=Discovered2026.0584.3255,300
AutoTTS (β=1.0)Base Model=Qwen3-4B, Type=Discovered2026.0583.5424,900
AutoTTS (β=0.5)Base Model=Qwen3-4B, Type=Discovered2026.0582236,700
Parallel-ProbeBase Model=Qwen3-8B, Type=Handcrafted2026.0581.5730,800
SC @ 64Base Model=Qwen3-8B, Type=Handcrafted2026.0580.4910,800
ASCBase Model=Qwen3-8B, Type=Handcrafted2026.0580.4226,000
ESCBase Model=Qwen3-8B, Type=Handcrafted2026.0580.4459,400
ASCBase Model=Qwen3-4B, Type=Handcrafted2026.0580.1175,700
SC @ 64Base Model=Qwen3-4B, Type=Handcrafted2026.0580886,800
ESCBase Model=Qwen3-4B, Type=Handcrafted2026.0580528,900
Parallel-ProbeBase Model=Qwen3-4B, Type=Handcrafted2026.0579.7688,900
SC @ 64Base Model=Qwen3-1.7B, Type=Handcrafted2026.0572.51,025,800
ESCBase Model=Qwen3-1.7B, Type=Handcrafted2026.0572.5909,200
ASCBase Model=Qwen3-1.7B, Type=Handcrafted2026.0572.3482,600
AutoTTS (β=1.0)Base Model=Qwen3-1.7B, Type=Discovered2026.0570.4499,100
AutoTTS (β=0.5)Base Model=Qwen3-1.7B, Type=Discovered2026.0568.5276,300
Parallel-ProbeBase Model=Qwen3-1.7B, Type=Handcrafted2026.0568.1748,500
Parallel-ProbeBase Model=Qwen3-0.6B, Type=Handcrafted2026.0521.8773,800
SC @ 64Base Model=Qwen3-0.6B, Type=Handcrafted2026.0521.41,008,600
ASCBase Model=Qwen3-0.6B, Type=Handcrafted2026.0521.4805,500
ESCBase Model=Qwen3-0.6B, Type=Handcrafted2026.0521.4986,700
AutoTTS (β=1.0)Base Model=Qwen3-0.6B, Type=Discovered2026.0520.9542,200
AutoTTS (β=0.5)Base Model=Qwen3-0.6B, Type=Discovered2026.0519.2283,600