List your company

Scientific Reasoning on GPQA (Acc, Tok, CR)

66.2AccuracyDTSR
29.2838.86548.4558.035Mar 13, 2026Mar 17, 2026Mar 21, 2026Mar 26, 2026Mar 30, 2026Apr 3, 2026Apr 8, 2026

Evaluation results

MethodLinks
DTSRBackbone=Qwen3-14B2026.0466.25,916—
VanillaBackbone=Qwen3-14B2026.0465.77,513—
TALEBackbone=Qwen3-14B2026.0465.67,138—
VanillaBackbone=Ministral-3-14B-Reasoning-25122026.0362.79,539100
VanillaBackbone=Ministral-3-8B-Reasoning-25122026.0362.311,765100
DEERBackbone=Ministral-3-8B-Reasoning-25122026.0361.111,31294.1
TERMINATORBackbone=Ministral-3-14B-Reasoning-25122026.0360.67,72776.5
VanillaBackbone=Qwen3-14B2026.0359.97,628100
DEERBackbone=Ministral-3-14B-Reasoning-25122026.0358.59,18595
TERMINATORBackbone=Ministral-3-8B-Reasoning-25122026.0358.29,58877.4
VanillaBackbone=Qwen3-8B2026.0357.68,594100
Thought-CalibBackbone=Qwen3-14B2026.0354.65,75781.9
TERMINATORBackbone=Qwen3-14B2026.0353.96,79887.1
Thought-CalibBackbone=Qwen3-8B2026.0352.66,24078.9
TERMINATORBackbone=Qwen3-8B2026.0352.17,54385.7
DynasorBackbone=Ministral-3-8B-Reasoning-25122026.0351.89,76683.7
DynasorBackbone=Ministral-3-14B-Reasoning-25122026.0351.37,91785.2
CoDBackbone=Qwen3-14B2026.04511,177—
DEERBackbone=Qwen3-8B2026.0350.58,55399.6
Thought-CalibBackbone=Ministral-3-14B-Reasoning-25122026.0349.56,02873.8
DEERBackbone=Qwen3-14B2026.03497,45197.4
Thought-CalibBackbone=Ministral-3-8B-Reasoning-25122026.0347.37,40671.8
DynasorBackbone=Qwen3-14B2026.0345.62,10129.1
DynasorBackbone=Qwen3-8B2026.0343.22,45528.4
NoThinkingBackbone=Ministral-3-14B-Reasoning-25122026.03415706.6
NoThinkingBackbone=Ministral-3-8B-Reasoning-25122026.0339.72,10616
NoThinkingBackbone=Qwen3-14B2026.0332.11,26518.8
NoThinkingBackbone=Qwen3-8B2026.0330.71,20415.8