Parsel: Algorithmic Reasoning with Language Models by Composing Decompositions

Understand

Despite recent success in large language model (LLM) reasoning, LLMs struggle with hierarchical multi-step reasoning tasks like generating complex programs.

• For these tasks, humans often start with a high-level algorithmic design and implement each part gradually.
• We introduce Parsel, a framework enabling automatic implementation and validation of complex algorithms with code LLMs.
• With Parsel, we automatically decompose algorithmic tasks into hierarchical natural language function descriptions and then search over combinations of possible function implementations using tests.