List your company

Explorative Reasoning on Crosswords Word-level (test)

13.5AccuracyCoT
-0.023.49710.51Mar 6, 2026

Evaluation results

MethodLinks
CoTBackbone=Qwen3-30B2026.0313.513,201
RouteGoTBackbone={Qwen3-4B, 8B, 30B}2026.03113,804
GoT*Backbone=Qwen3-30B2026.036.511,674
IOBackbone=Qwen3-30B2026.036467
EmbedLLMBackbone={Qwen3-4B, 8B, 30B}2026.0356,017
AGoTBackbone=Qwen3-30B2026.033.520,893
RouteLLMBackbone={Qwen3-4B, 8B, 30B}2026.032.55,831
ToTBackbone=Qwen3-30B2026.032.19,563
RTRBackbone={Qwen3-4B, 8B, 30B}2026.031.54,065
RandomBackbone={Qwen3-4B, 8B, 30B}2026.0316,635
KNNBackbone={Qwen3-4B, 8B, 30B}2026.030.55,813