RLMOpt: Adaptive Prompt Optimization via Recursive Language Models

Subhash Bangalore Satheesha, Nirvik Pande, Deepthi Duddempudi +1

Prompt optimizers automate the search for prompts that improve language-model performance, but existing methods rely on a predefined optimization procedure: the algorithm determine…