Question Answering on OBQA (Accuracy)
86AccuracyExpGraph
Evaluation results
| Method | Links | |
|---|---|---|
| ExpGraphBackbone Model=Llama-3.1-8B-Instruct (Large LLM), Method Category=LLM-Centric Experience Learning Baselines2026.05 | 86 | |
| ExpGraphBackbone Model=Llama-3.2-3B-Instruct (Small LLM), Method Category=LLM-Centric Experience Learning Baselines2026.05 | 74 | |
| FP16Backbone=Qwen-3-8B, Evaluation Protocol=zero-shot2026.06 | 42.4 | |
| LLM-PrunerBackbone=Qwen2.5-32B-Instruct, Sparsity=50%2026.06 | 42 | |
| Persona-PrunerBackbone=Qwen2.5-32B-Instruct, Sparsity=50%2026.06 | 42 | |
| SAGE-PTQBackbone=Qwen-3-8B, Evaluation Protocol=zero-shot2026.06 | 41 | |
| FP16Backbone=LLaMA-3-8B, Evaluation Protocol=zero-shot2026.06 | 40.6 | |
| SAGE-PTQBackbone=LLaMA-3-8B, Evaluation Protocol=zero-shot2026.06 | 38.2 | |
| BiLLMBackbone=Qwen-3-8B, Evaluation Protocol=zero-shot2026.06 | 37.1 | |
| Adapt-PrunerBackbone=Qwen2.5-32B-Instruct, Sparsity=50%2026.06 | 37 | |
| FP16Backbone=OPT-6.7B, Evaluation Protocol=zero-shot2026.06 | 31.6 | |
| BiLLMBackbone=LLaMA-3-8B, Evaluation Protocol=zero-shot2026.06 | 30.14 | |
| Depth PruningBackbone=Qwen2.5-32B-Instruct, Sparsity=50%2026.06 | 30 | |
| SliceGPTBackbone=Qwen2.5-32B-Instruct, Sparsity=50%2026.06 | 30 | |
| SAGE-PTQBackbone=OPT-6.7B, Evaluation Protocol=zero-shot2026.06 | 29.2 | |
| BiLLMBackbone=OPT-6.7B, Evaluation Protocol=zero-shot2026.06 | 29 |