List your company

Mathematical Reasoning on AQuA (Verifier Settings)

74Accuracy (Without Verifier)ToT
19.9233.964862.04Mar 28, 2025

Evaluation results

MethodLinks
ToTModel=Llama-3.1-70B2025.03747474
CoTModel=Llama-3.1-70B2025.03727373
L2MModel=Llama-3.1-70B2025.03727273
MCTSModel=Llama-3.1-70B2025.03727372
CoTModel=Llama-3.1-8B2025.03606360
L2MModel=Llama-3.1-8B2025.03586258
ToTModel=Llama-3.1-8B2025.03505350
MCTSModel=Llama-3.1-8B2025.03505150
CoTModel=Llama-3.2-3B2025.03465146
ToTModel=Llama-3.2-1B2025.03353835
MCTSModel=Llama-3.2-3B2025.03353635
ToTModel=Llama-3.2-3B2025.03333533
MCTSModel=Llama-3.2-1B2025.03293231
L2MModel=Llama-3.2-3B2025.03293131
CoTModel=Llama-3.2-1B2025.03262826
L2MModel=Llama-3.2-1B2025.03222429