List your company

Spoken Mathematics Reasoning on VoxEval High School 1.0 (test)

81.9AccuracyTGS
18.66835.08451.567.916Mar 15, 2026

Evaluation results

MethodLinks
TGSModel=Qwen2.5, Hyperparams. (k, α)=3, 0.052026.0381.9
SGSModel=Qwen2.5, Hyperparams. (k, α)=4, 0.0252026.0380.7
CoTModel=Qwen2.5, Hyperparams. (k, α)=-2026.0379.9
VanillaModel=Qwen2.5, Hyperparams. (k, α)=3, 0.0252026.0379.6
TGSModel=Phi4-mm, Hyperparams. (k, α)=3, 0.0252026.0348.9
VanillaModel=Phi4-mm, Hyperparams. (k, α)=5, 0.0252026.0347
SGSModel=Phi4-mm, Hyperparams. (k, α)=3, 0.052026.0347
NormalModel=Qwen2.5, Hyperparams. (k, α)=-2026.0345.6
SGSModel=Voxtral, Hyperparams. (k, α)=3, 0.12026.0343.9
VanillaModel=Voxtral, Hyperparams. (k, α)=5, 0.0252026.0343.3
CoTModel=Voxtral, Hyperparams. (k, α)=-2026.0342
CoTModel=Phi4-mm, Hyperparams. (k, α)=-2026.0341.9
TGSModel=Voxtral, Hyperparams. (k, α)=4, 0.0252026.0341.3
NormalModel=Voxtral, Hyperparams. (k, α)=-2026.0336.7
NormalModel=AF3, Hyperparams. (k, α)=-2026.0328.2
NormalModel=Phi4-mm, Hyperparams. (k, α)=-2026.0324.8
CoTModel=AF3, Hyperparams. (k, α)=-2026.0324.2
TGSModel=AF3, Hyperparams. (k, α)=5, 0.12026.0324.1
VanillaModel=AF3, Hyperparams. (k, α)=3, 0.12026.0322.2
SGSModel=AF3, Hyperparams. (k, α)=1, 0.0252026.0321.1