Reading Comprehension on BoolQ (dev)
84.8AccuracyMegatron-NLG
Evaluation results
| Method | Links | |
|---|---|---|
| Megatron-NLGModel Size=530B, Evaluation Protocol=Few-shot2021.12 | 84.8 | |
| GLaMModel Size=64B/64E, Evaluation Protocol=Zero-shot2021.12 | 83.1 | |
| GLaMModel Size=64B/64E, Evaluation Protocol=Few-shot, Shots=82021.12 | 83.1 | |
| GLaMModel Size=64B/64E, Evaluation Protocol=One-shot2021.12 | 82.8 | |
| GPT-3Model Size=175B, Evaluation Protocol=Few-shot, Shots=322021.12 | 77.5 | |
| GPT-3Model Size=175B, Evaluation Protocol=One-shot2021.12 | 76.7 | |
| Hier-GateNetDiscovery Source=AI Discovered2026.03 | 63.03 | |
| Content-SharpRouterDiscovery Source=AI Discovered2026.03 | 62.51 | |
| PathGate-FusionNetDiscovery Source=AI Discovered2026.03 | 62.45 | |
| FusionGated-FIRNetDiscovery Source=AI Discovered2026.03 | 62.11 | |
| DeltaNetDiscovery Source=Human Discovered2026.03 | 60.58 | |
| Mamba2Discovery Source=Human Discovered2026.03 | 60.52 | |
| GPT-3Model Size=175B, Evaluation Protocol=Zero-shot2021.12 | 60.5 | |
| AdaMulti-PathGateNetDiscovery Source=AI Discovered2026.03 | 56.27 | |
| Gated DeltaNetDiscovery Source=Human Discovered2026.03 | 53.98 |