NLU Question Answering on PIQA (Accuracy)
80.25PIQA AccuracyFP16
Evaluation Results
| Method | Links | |
|---|---|---|
| FP16Backbone=LLaMA-3-8B, Evaluation Protocol=zero-shot2026.06 | 80.25 | |
| FP16Backbone=Qwen-3-8B, Evaluation Protocol=zero-shot2026.06 | 78.94 | |
| SAGE-PTQBackbone=LLaMA-3-8B, Evaluation Protocol=zero-shot2026.06 | 78.78 | |
| SAGE-PTQBackbone=Qwen-3-8B, Evaluation Protocol=zero-shot2026.06 | 77.31 | |
| FP16Backbone=OPT-6.7B, Evaluation Protocol=zero-shot2026.06 | 75.68 | |
| SAGE-PTQBackbone=OPT-6.7B, Evaluation Protocol=zero-shot2026.06 | 73.94 | |
| BiLLMBackbone=Qwen-3-8B, Evaluation Protocol=zero-shot2026.06 | 62.5 | |
| BiLLMBackbone=OPT-6.7B, Evaluation Protocol=zero-shot2026.06 | 58.6 | |
| BiLLMBackbone=LLaMA-3-8B, Evaluation Protocol=zero-shot2026.06 | 57.51 |