Med-Diet significantly outperformed GPT-4o ( p < 0.05) and Llama-3.3-70B ( p < 0.05), but its differences from DeepSeek-R1 ( p > 0.05) and GLM-Z1-32B ( p > 0.05) did not reach statistical significance.
← all excerpts
Med-Diet: evaluation of an LLM-based system for clinically guided nutrition care in chronic diseases.
1
—
—