List your company
ResearchFormerly WizWand

Reasoning on AIME 25 (Accuracy, Consistency, # traces)

681Trace CountPETS-On. (conf)
174.52306.01437.5568.99Feb 18, 2026
Updated 8mo ago

Evaluation results

MethodLinks
PETS-On. (conf)LLM=Qwen-Long, Confidence-Weighted=true2026.026819685.7
PETS-On. (conf)LLM=Qwen3-30B, Confidence-Weighted=true2026.025359782.5
PETS-On.LLM=Qwen3-30B, Confidence-Weighted=false2026.0250410083.5
Uniform (conf)LLM=Qwen-Long, Confidence-Weighted=true2026.024959383.5
PETS-On. (conf)LLM=GPT-120B, Confidence-Weighted=true2026.024939892.3
Uniform (conf)LLM=Qwen3-30B, Confidence-Weighted=true2026.024869481.2
Uniform (conf)LLM=GPT-120B, Confidence-Weighted=true2026.024599489.8
PETS-On.LLM=GPT-120B, Confidence-Weighted=false2026.0245410091.7
Uniform (conf)LLM=Qwen3-4B, Confidence-Weighted=true2026.024529276.3
UniformLLM=GPT-120B, Confidence-Weighted=false2026.024169589.2
UniformLLM=Qwen3-30B, Confidence-Weighted=false2026.024159581
Uniform (conf)LLM=GPT-20B, Confidence-Weighted=true2026.023499385.2
PETS-On. (conf)LLM=GPT-20B, Confidence-Weighted=true2026.023309684.8
UniformLLM=Qwen3-4B, Confidence-Weighted=false2026.022909574.8
PETS-On. (conf)LLM=Qwen3-4B, Confidence-Weighted=true2026.022879576.3
UniformLLM=Qwen-Long, Confidence-Weighted=false2026.022299584.7
UniformLLM=GPT-20B, Confidence-Weighted=false2026.022219686
PETS-On.LLM=Qwen-Long, Confidence-Weighted=false2026.0221810085
PETS-On.LLM=GPT-20B, Confidence-Weighted=false2026.0219510085
PETS-On.LLM=Qwen3-4B, Confidence-Weighted=false2026.0219410075