List your company

Spoken Scientific Reasoning on ReveAL-CoT SpeechR 1.0 (test)

74AccuracyVanilla
37.0846.66556.2565.835Mar 15, 2026

Evaluation results

MethodLinks
VanillaModel=Qwen2.5, Hyperparams. (k, α)=3, 0.0252026.0374
SGSModel=Qwen2.5, Hyperparams. (k, α)=4, 0.0252026.0371.5
TGSModel=Qwen2.5, Hyperparams. (k, α)=3, 0.052026.0371.5
CoTModel=Qwen2.5, Hyperparams. (k, α)=-2026.0371
SGSModel=Phi4-mm, Hyperparams. (k, α)=3, 0.052026.0370.5
TGSModel=Phi4-mm, Hyperparams. (k, α)=3, 0.0252026.0370.5
VanillaModel=Phi4-mm, Hyperparams. (k, α)=5, 0.0252026.0369.5
CoTModel=Phi4-mm, Hyperparams. (k, α)=-2026.0368.5
SGSModel=Voxtral, Hyperparams. (k, α)=3, 0.12026.0357
VanillaModel=Voxtral, Hyperparams. (k, α)=5, 0.0252026.0356
TGSModel=Voxtral, Hyperparams. (k, α)=4, 0.0252026.0354.5
NormalModel=Voxtral, Hyperparams. (k, α)=-2026.0353.5
NormalModel=Qwen2.5, Hyperparams. (k, α)=-2026.0351.5
VanillaModel=AF3, Hyperparams. (k, α)=3, 0.12026.0347.5
NormalModel=AF3, Hyperparams. (k, α)=-2026.0346.5
TGSModel=AF3, Hyperparams. (k, α)=5, 0.12026.0346.5
CoTModel=AF3, Hyperparams. (k, α)=-2026.0345.5
CoTModel=Voxtral, Hyperparams. (k, α)=-2026.0343
SGSModel=AF3, Hyperparams. (k, α)=1, 0.0252026.0342.5
NormalModel=Phi4-mm, Hyperparams. (k, α)=-2026.0338.5