Open-ended Question Answering on HotpotQA (test)
88AccuracyRouteGoT
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| RouteGoTBackbone={Qwen3-4B, 8B, 30B}2026.03 | 88 | 592 | |
| GoT*Backbone=Qwen3-30B2026.03 | 87 | 812 | |
| CoTBackbone=Qwen3-30B2026.03 | 86 | 431 | |
| RouteLLMBackbone={Qwen3-4B, 8B, 30B}2026.03 | 83 | 555 | |
| IOBackbone=Qwen3-30B2026.03 | 81 | 11 | |
| KNNBackbone={Qwen3-4B, 8B, 30B}2026.03 | 79 | 627 | |
| RTRBackbone={Qwen3-4B, 8B, 30B}2026.03 | 79 | 554 | |
| RandomBackbone={Qwen3-4B, 8B, 30B}2026.03 | 77 | 1,685 | |
| EmbedLLMBackbone={Qwen3-4B, 8B, 30B}2026.03 | 77 | 2,040 | |
| AGoTBackbone=Qwen3-30B2026.03 | 72 | 2,583 | |
| ToTBackbone=Qwen3-30B2026.03 | 60 | 3,748 |