List your company

Mathematical Reasoning on MATH500 (Acc, Tp, Spd)

83.2AccuracyCSD
43.88854.09464.374.506Apr 15, 2026

Evaluation results

MethodLinks
CSDModel Series=Qwen-2.5 (72B/7B), Shot=4-shot, Protocol=Pass@12026.0483.232.82.12
Vanilla DecodingModel Series=Qwen-2.5 (72B/7B), Shot=4-shot, Protocol=Pass@12026.0482.215.41
SpecDecodeModel Series=Qwen-2.5 (72B/7B), Shot=4-shot, Protocol=Pass@12026.0481.429.61.92
Lossy SDModel Series=Llama-3 (70B/1B), Shot=4-shot, Protocol=Pass@1, τ=0.62026.0449.430.62
CSDModel Series=Llama-3 (70B/1B), Shot=4-shot, Protocol=Pass@12026.044835.52.33
LookaheadModel Series=Llama-3 (70B/1B), Shot=4-shot, Protocol=Pass@12026.0446.612.50.82
SWIFTModel Series=Llama-3 (70B/1B), Shot=4-shot, Protocol=Pass@12026.0446.416.51.09
Vanilla DecodingModel Series=Llama-3 (70B/1B), Shot=4-shot, Protocol=Pass@12026.044615.21
SpecDecodeModel Series=Llama-3 (70B/1B), Shot=4-shot, Protocol=Pass@12026.0445.428.81.89