List your company

Question Answering on OpenBookQA (Accuracy, TFLOPs)

93.6Accuracydense alignment
-1.45623.22247.972.578Jun 11, 2026

Evaluation results

MethodLinks
dense alignmentPair=4B → 14B, Regime=context-aware2026.0693.621.54
dense alignmentPair=8B → 14B, Regime=context-aware2026.0693.421.79
T2TPair=8B → 14B, Regime=context-aware2026.0692.867.08
dense alignmentPair=14B → 8B, Regime=context-aware2026.0692.615.64
T2TPair=4B → 14B, Regime=context-aware2026.069256.24
Receiver-onlyPair=4B → 14B, Regime=context-aware2026.0691.831.16
Receiver-onlyPair=8B → 14B, Regime=context-aware2026.0691.831.16
Receiver-onlyPair=4B → 8B, Regime=context-aware2026.0691.219.85
dense alignmentPair=4B → 8B, Regime=context-aware2026.0691.212.56
Receiver-onlyPair=14B → 8B, Regime=context-aware2026.0691.219.85
T2TPair=4B → 8B, Regime=context-aware2026.0690.437.73
T2TPair=14B → 8B, Regime=context-aware2026.0690.255.5
Ours-context-unawarePair=8B → 4B, Context Setting=context-unaware2026.06906.56
dense alignmentPair=8B → 4B, Regime=context-aware2026.0689.67.95
Ours-context-unawarePair=4B → 8B, Context Setting=context-unaware2026.0688.89.42
dense alignmentPair=14B → 4B, Regime=context-aware2026.0688.810.18
Ours-context-unawarePair=14B → 8B, Context Setting=context-unaware2026.068813.28
Receiver-onlyPair=8B → 4B, Regime=context-aware2026.06889.18
Receiver-onlyPair=14B → 4B, Regime=context-aware2026.06889.18
C2CPair=4B → 14B, Regime=context-aware2026.06876.64
T2TPair=8B → 4B, Regime=context-aware2026.068733.91
C2CPair=8B → 14B, Regime=context-aware2026.06877.42
C2CPair=4B → 8B, Regime=context-aware2026.0685.64.5
C2CPair=14B → 8B, Regime=context-aware2026.0685.67.83
Ours-context-unawarePair=14B → 4B, Context Setting=context-unaware2026.0685.48.9
T2TPair=14B → 4B, Regime=context-aware2026.068543.48
Ours-context-unawarePair=8B → 14B, Context Setting=context-unaware2026.0683.815.85
Ours-context-unawarePair=4B → 14B, Context Setting=context-unaware2026.0682.417.15
C2CPair=8B → 4B, Regime=context-aware2026.0677.23.43
C2CPair=14B → 4B, Regime=context-aware2026.0676.45.01
T2T-context-unawarePair=8B → 14B, Context Setting=context-unaware2026.0627.472.06
T2T-context-unawarePair=4B → 14B, Context Setting=context-unaware2026.0627.260.83
C2C-context-unawarePair=4B → 8B, Context Setting=context-unaware2026.062714.94
T2T-context-unawarePair=14B → 8B, Context Setting=context-unaware2026.0626.253.8
T2T-context-unawarePair=4B → 8B, Context Setting=context-unaware2026.062437.86
T2T-context-unawarePair=8B → 4B, Context Setting=context-unaware2026.0623.232.75
T2T-context-unawarePair=14B → 4B, Context Setting=context-unaware2026.0620.241.95
C2C-context-unawarePair=4B → 14B, Context Setting=context-unaware2026.0616.420.05
C2C-context-unawarePair=8B → 4B, Context Setting=context-unaware2026.061014.76
C2C-context-unawarePair=8B → 14B, Context Setting=context-unaware2026.067.69.84
C2C-context-unawarePair=14B → 4B, Context Setting=context-unaware2026.066.617.93
C2C-context-unawarePair=14B → 8B, Context Setting=context-unaware2026.062.224.77