Science Question Answering on ARC Easy (Accuracy & Exit Position)
86.5AccuracyMinistral3 8B
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Ministral3 8BModel Architecture=Ministral3 8B (34 layers), Evaluation Protocol=Backbone2026.04 | 86.5 | — | |
| River-LLMModel Architecture=Ministral3 8B (34 layers), Early-exit threshold (tau)=0.8, Evaluation Protocol=Adaptive2026.04 | 85.9 | 2.01 | |
| River-LLMModel Architecture=Ministral3 8B (34 layers), Early-exit threshold (tau)=0.9, Evaluation Protocol=Adaptive2026.04 | 85.9 | 2.11 | |
| Phi4-miniModel Architecture=Phi4-mini (32 layers), Evaluation Protocol=Backbone2026.04 | 82.8 | — | |
| River-LLMModel Architecture=Phi4-mini (32 layers), Early-exit threshold (tau)=0.8, Evaluation Protocol=Adaptive2026.04 | 81.3 | 2.02 | |
| River-LLMModel Architecture=Phi4-mini (32 layers), Early-exit threshold (tau)=0.9, Evaluation Protocol=Adaptive2026.04 | 81.3 | 3.45 |