List your company

Reasoning on Reasoning Domain Code

76.25Reasoning ScoreSTEER2ADAPT
48.22255.498562.77570.0515Feb 7, 2026

Evaluation results

MethodLinks
STEER2ADAPTBackbone=Qwen-2.5-7B-Instruct2026.0276.25
Few-ShotBackbone=Qwen-2.5-7B-Instruct, n=22026.0272.54
REPBackbone=Qwen-2.5-7B-Instruct2026.0272.41
STEER2ADAPTBackbone=Llama-3.1-8B-Instruct2026.0272.25
CAABackbone=Qwen-2.5-7B-Instruct2026.0271.97
Few-ShotBackbone=Qwen-2.5-7B-Instruct, n=12026.0271.69
Direct InferenceBackbone=Qwen-2.5-7B-Instruct2026.0271.15
ICLBackbone=Qwen-2.5-7B-Instruct2026.0271.12
REPBackbone=Llama-3.1-8B-Instruct2026.0267
CAABackbone=Llama-3.1-8B-Instruct2026.0260.81
Direct InferenceBackbone=Llama-3.1-8B-Instruct2026.0259.11
ICLBackbone=Llama-3.1-8B-Instruct2026.0257.93
Few-ShotBackbone=Llama-3.1-8B-Instruct, n=12026.0255.47
Few-ShotBackbone=Llama-3.1-8B-Instruct, n=22026.0255.23
STEER2ADAPTBackbone=Mistral-7B-Instruct-v0.12026.0252.65
REPBackbone=Mistral-7B-Instruct-v0.12026.0251.4
Few-ShotBackbone=Mistral-7B-Instruct-v0.1, n=12026.0249.69
Few-ShotBackbone=Mistral-7B-Instruct-v0.1, n=22026.0249.69
ICLBackbone=Mistral-7B-Instruct-v0.12026.0249.65
Direct InferenceBackbone=Mistral-7B-Instruct-v0.12026.0249.49
CAABackbone=Mistral-7B-Instruct-v0.12026.0249.3