List your company

Logical Reasoning on Reasoning Gym Zebra Puzzles

39.33Accuracy (*)TokUR (TU)
21.306825.985930.66535.3441May 16, 2025

Evaluation results

MethodLinks
TokUR (TU)Base Model=Qwen-2.5-3B-Instruct2025.0539.3371.3841.42
TokUR (AU)Base Model=Qwen-2.5-3B-Instruct2025.0539.3371.6641.71
TokUR (EU)Base Model=Qwen-2.5-3B-Instruct2025.0537.3371.2841.2
CoT (Lower-Bound)Base Model=Qwen-2.5-3B-Instruct2025.0533.67——
Self-CertaintyBase Model=Qwen-2.5-3B-Instruct2025.053047.7722.95
DeepConfBase Model=Qwen-2.5-3B-Instruct2025.052642.4121.07
PEBase Model=Qwen-2.5-3B-Instruct2025.052444.721.76
LLBase Model=Qwen-2.5-3B-Instruct2025.052242.1521.02