Explorative Reasoning on Crosswords Letter-level (test)
23.4AccuracyRouteGoT
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| RouteGoTBackbone={Qwen3-4B, 8B, 30B}2026.03 | 23.4 | 3,804 | |
| GoT*Backbone=Qwen3-30B2026.03 | 22.4 | 11,674 | |
| CoTBackbone=Qwen3-30B2026.03 | 22.2 | 13,201 | |
| IOBackbone=Qwen3-30B2026.03 | 22 | 467 | |
| EmbedLLMBackbone={Qwen3-4B, 8B, 30B}2026.03 | 19 | 6,017 | |
| RouteLLMBackbone={Qwen3-4B, 8B, 30B}2026.03 | 12.8 | 5,831 | |
| ToTBackbone=Qwen3-30B2026.03 | 12.4 | 9,563 | |
| AGoTBackbone=Qwen3-30B2026.03 | 11 | 20,893 | |
| RTRBackbone={Qwen3-4B, 8B, 30B}2026.03 | 9.4 | 4,065 | |
| RandomBackbone={Qwen3-4B, 8B, 30B}2026.03 | 7.6 | 6,635 | |
| KNNBackbone={Qwen3-4B, 8B, 30B}2026.03 | 6.2 | 5,813 |