List your company
ResearchFormerly WizWand

Question Answering on GPQA (Accuracy, Speedup)

31.03AccuracyProphet
25.923627.249328.57529.9007Dec 2, 2025
Updated 8mo ago

Evaluation results

MethodLinks
ProphetBackbone=LLaDA Base2025.1231.0315.32
Exp-k=16 (7.5, 2.5)Scheduling Strategy=Exponential (SchED), Curvature (k)=16, Thresholds (tau_high, tau_low)=(7.5, 2.5), Backbone=LLaDA Base2025.1230.84.47
Cosine (7.5, 0)Schedule Type=Cosine, Thresholds (τ_high, τ_low)=(7.5, 0)2025.1230.5815.96
Cosine (7.5, 2.5)Schedule Type=Cosine, Thresholds (τ_high, τ_low)=(7.5, 2.5)2025.1230.5815.93
Linear (7.5, 0)Schedule Type=Linear, Thresholds (τ_high, τ_low)=(7.5, 0)2025.1230.5816.08
Linear (7.5, 2.5)Schedule Type=Linear, Thresholds (τ_high, τ_low)=(7.5, 2.5)2025.1230.5816.01
Exp-k=2 (7.5, 0)Schedule Type=Exponential, k=2, Thresholds (τ_high, τ_low)=(7.5, 0)2025.1230.5816.32
Exp-k=16 (7.5, 2.5)Schedule Type=Exponential, k=16, Thresholds (τ_high, τ_low)=(7.5, 2.5)2025.1230.5817.16
BaselineSchedule=Standard2025.1230.361
Exp-k=2 (7.5, 2.5)Schedule Type=Exponential, k=2, Thresholds (τ_high, τ_low)=(7.5, 2.5)2025.1230.1316.18
Exp-k=16 (7.5, 0)Schedule Type=Exponential, k=16, Thresholds (τ_high, τ_low)=(7.5, 0)2025.1229.4618.64
Exp-k=2 (7.5, 0)Scheduling Strategy=Exponential (SchED), Curvature (k)=2, Thresholds (tau_high, tau_low)=(7.5, 0), Backbone=LLaDA Base2025.1229.022.85
ProphetSchedule=Prophet2025.1228.7916.45
Exp-k=16 (7.5, 0)Scheduling Strategy=Exponential (SchED), Curvature (k)=16, Thresholds (tau_high, tau_low)=(7.5, 0), Backbone=LLaDA Base2025.1228.798.67
Exp-k=4 (7.5, 2.5)Scheduling Strategy=Exponential (SchED), Curvature (k)=4, Thresholds (tau_high, tau_low)=(7.5, 2.5), Backbone=LLaDA Base2025.1228.792.86
Exp-k=8 (7.5, 2.5)Scheduling Strategy=Exponential (SchED), Curvature (k)=8, Thresholds (tau_high, tau_low)=(7.5, 2.5), Backbone=LLaDA Base2025.1228.573.7
Linear (7.5, 0)Scheduling Strategy=Linear, Thresholds (tau_high, tau_low)=(7.5, 0), Backbone=LLaDA Base2025.1228.572.35
Exp-k=8 (7.5, 0)Scheduling Strategy=Exponential (SchED), Curvature (k)=8, Thresholds (tau_high, tau_low)=(7.5, 0), Backbone=LLaDA Base2025.1228.355.94
Linear (7.5, 2.5)Scheduling Strategy=Linear, Thresholds (tau_high, tau_low)=(7.5, 2.5), Backbone=LLaDA Base2025.1228.121.98
Cosine (7.5, 0)Scheduling Strategy=Cosine, Thresholds (tau_high, tau_low)=(7.5, 0), Backbone=LLaDA Base2025.1227.92.24
Exp-k=4 (7.5, 0)Scheduling Strategy=Exponential (SchED), Curvature (k)=4, Thresholds (tau_high, tau_low)=(7.5, 0), Backbone=LLaDA Base2025.1227.464.09
Cosine (7.5, 2.5)Scheduling Strategy=Cosine, Thresholds (tau_high, tau_low)=(7.5, 2.5), Backbone=LLaDA Base2025.1227.231.97
Exp-k=2 (7.5, 2.5)Scheduling Strategy=Exponential (SchED), Curvature (k)=2, Thresholds (tau_high, tau_low)=(7.5, 2.5), Backbone=LLaDA Base2025.1226.342.24
BaselineBackbone=LLaDA Base2025.1226.121