TextGrad: Automatic "Differentiation" via Text

1 code implementation • 11 Jun 2024 • Mert Yuksekgonul, Federico Bianchi, Joseph Boen, Sheng Liu, Zhi Huang, Carlos Guestrin, James Zou

Without modifying the framework, TextGrad improves the zero-shot accuracy of GPT-4o in Google-Proof Question Answering from $51\%$ to $55\%$, yields $20\%$ relative performance gain in optimizing LeetCode-Hard coding problem solutions, improves prompts for reasoning, designs new druglike small molecules with desirable in silico binding, and designs radiation oncology treatment plans with high specificity.

Question Answering Specificity