Online Prompt Selection for Program Synthesis

TL;DR

This paper introduces CYANEA, an online learning system that dynamically selects the best solver or prompt combination for program synthesis tasks, significantly improving success rates and efficiency over static approaches.

Contribution

It presents a novel multi-armed bandit based method for adaptive solver and prompt selection in program synthesis, addressing non-expert user challenges.

Findings

CYANEA solves 37.2% more queries than the best single solver.

Achieves results within 4% of the virtual best solver.

Effective across various synthesis benchmarks.

Abstract

Large Language Models (LLMs) demonstrate impressive capabilities in the domain of program synthesis. This level of performance is not, however, universal across all tasks, all LLMs and all prompting styles. There are many areas where one LLM dominates, one prompting style dominates, or where calling a symbolic solver is a better choice than an LLM. A key challenge for the user then, is to identify not only when an LLM is the right choice of solver, and the appropriate LLM to call for a given synthesis task, but also the right way to call it. A non-expert user who makes the wrong choice, incurs a cost both in terms of results (number of tasks solved, and the time it takes to solve them) and financial cost, if using a closed-source language model via a commercial API. We frame this choice as an online learning problem. We use a multi-armed bandit algorithm to select which symbolic solver, or LLM and prompt combination to deploy in order to maximize a given reward function (which may prioritize solving time, number of synthesis tasks solved, or financial cost of solving). We implement an instance of this approach, called CYANEA, and evaluate it on synthesis queries from the literature in ranking function synthesis, from the syntax-guided synthesis competition, and fresh, unseen queries generated from SMT problems. CYANEA solves 37.2% more queries than the best single solver and achieves results within 4% of the virtual best solver.