List your company

Mathematical Reasoning on GSM8K (Source Lib., Unseen Lib., Delta)

93.21Accuracy (Source Lib.)CARD
87.084488.674790.26591.8553Jun 18, 2026

Evaluation results

MethodLinks
CARDBase Model=GPT-4o-mini2026.0693.2191.441.77
SIGMABase Model=GPT-4o-mini2026.0692.9691.861.1
G-DesignerBase Model=GPT-4o-mini2026.0689.2687.281.98
GPTSwarmBase Model=GPT-4o-mini2026.0687.8384.473.36
LLM-DebateBase Model=GPT-4o-mini2026.0687.3285.292.03