List your company

Reasoning on Reasoning Domain Logic

79.8ScoreREP
48.485656.615364.74572.8747Feb 7, 2026

Evaluation results

MethodLinks
REPBackbone=Qwen-2.5-7B-Instruct2026.0279.8
STEER2ADAPTBackbone=Qwen-2.5-7B-Instruct2026.0279.68
Direct InferenceBackbone=Qwen-2.5-7B-Instruct2026.0279.45
CAABackbone=Qwen-2.5-7B-Instruct2026.0277.07
Few-ShotBackbone=Qwen-2.5-7B-Instruct, n=12026.0275.44
ICLBackbone=Qwen-2.5-7B-Instruct2026.0274.56
Few-ShotBackbone=Qwen-2.5-7B-Instruct, n=22026.0274.49
STEER2ADAPTBackbone=Llama-3.1-8B-Instruct2026.0269.27
STEER2ADAPTBackbone=Mistral-7B-Instruct-v0.12026.0267.89
Direct InferenceBackbone=Mistral-7B-Instruct-v0.12026.0266.9
Direct InferenceBackbone=Llama-3.1-8B-Instruct2026.0264.57
Few-ShotBackbone=Llama-3.1-8B-Instruct, n=12026.0264.37
CAABackbone=Llama-3.1-8B-Instruct2026.0263.7
Few-ShotBackbone=Llama-3.1-8B-Instruct, n=22026.0262.89
CAABackbone=Mistral-7B-Instruct-v0.12026.0262.35
REPBackbone=Mistral-7B-Instruct-v0.12026.0261.26
REPBackbone=Llama-3.1-8B-Instruct2026.0260.97
ICLBackbone=Mistral-7B-Instruct-v0.12026.0260.5
ICLBackbone=Llama-3.1-8B-Instruct2026.0259.48
Few-ShotBackbone=Mistral-7B-Instruct-v0.1, n=12026.0252.81
Few-ShotBackbone=Mistral-7B-Instruct-v0.1, n=22026.0249.69