Open-ended Question Answering on MoreHopQA (test)
77AccuracyRouteGoT
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| RouteGoTBackbone={Qwen3-4B, 8B, 30B}2026.03 | 77 | 665 | |
| GoT*Backbone=Qwen3-30B2026.03 | 73 | 814 | |
| RTRBackbone={Qwen3-4B, 8B, 30B}2026.03 | 73 | 626 | |
| KNNBackbone={Qwen3-4B, 8B, 30B}2026.03 | 71 | 624 | |
| RouteLLMBackbone={Qwen3-4B, 8B, 30B}2026.03 | 71 | 688 | |
| AGoTBackbone=Qwen3-30B2026.03 | 70 | 2,064 | |
| CoTBackbone=Qwen3-30B2026.03 | 68 | 727 | |
| EmbedLLMBackbone={Qwen3-4B, 8B, 30B}2026.03 | 65 | 2,621 | |
| RandomBackbone={Qwen3-4B, 8B, 30B}2026.03 | 64 | 1,516 | |
| ToTBackbone=Qwen3-30B2026.03 | 61 | 3,860 | |
| IOBackbone=Qwen3-30B2026.03 | 26 | 9 |