We Should Take Text Optimization More Seriously
On where learning should happen.
Yoonho Lee
Yoonho Lee is a final-year Stanford CS PhD student advised by Chelsea Finn, working on continual learning in text space, text optimization, AI agents, and LLM systems.
Latest posts
Meta-Harness: End-to-End Optimization of Model Harnesses
Project page for Meta-Harness. Interactive demo, results on text classification, math reasoning, and agentic coding.
What Is Taste?
Surveying attempts to define research taste, and why it matters for AI.
Are We Managers Now?
How the "maker's schedule" is changing with AI agents.
Following the Text Gradient at Scale
Feedback Descent scales text-based optimization into a simple, domain-agnostic procedure that outperforms specialized RL methods across molecular design, prompt optimization, and visual editing.