List your company

Inference Attack on GSM8K

0.9873AUCCIA
0.3874280.5431640.69890.854636Apr 14, 2026

Evaluation results

MethodLinks
CIAMAS Framework=ARG-Designer2026.040.987390.3583.3
CIAMAS Framework=AGP2026.040.880482.0478.66
CIAMAS Framework=G-Designer2026.040.858583.8277.44
Gemini-2.5-ProMAS Framework=ARG-Designer2026.040.747579.7757.43
GPT-5MAS Framework=ARG-Designer2026.040.698476.7251.13
GPT-5MAS Framework=G-Designer2026.040.627458.8745.55
GPT-5MAS Framework=AGP2026.040.623359.7253.57
Gemini-2.5-ProMAS Framework=AGP2026.040.617757.3351.74
Gemini-2.5-ProMAS Framework=G-Designer2026.040.615257.2845.45
Mistral-7B-Instruct-v0.2MAS Framework=ARG-Designer2026.040.571366.9749.24
Llama-3.1-8B-InstructMAS Framework=G-Designer2026.040.528453.4846.89
Llama-3.1-8B-InstructMAS Framework=ARG-Designer2026.040.512373.4564.35
Mistral-7B-Instruct-v0.2MAS Framework=AGP2026.040.48676528.66
Llama-3.1-8B-InstructMAS Framework=AGP2026.040.426551.951
Mistral-7B-Instruct-v0.2MAS Framework=G-Designer2026.040.410542.628.66