Minimization of Boolean Complexity in In-Context Concept Learning

TL;DR

This paper investigates how Boolean complexity influences in-context learning success in Large Language Models, revealing a bias towards simpler concepts similar to human learning patterns.

Contribution

It introduces a novel analysis linking Boolean complexity to LLM in-context learning performance, inspired by human concept learning insights.

Findings

Performance correlates with Boolean complexity

LLMs prefer simpler concepts in in-context learning

In-context learning exhibits a bias towards simplicity

Abstract

What factors contribute to the relative success and corresponding difficulties of in-context learning for Large Language Models (LLMs)? Drawing on insights from the literature on human concept learning, we test LLMs on carefully designed concept learning tasks, and show that task performance highly correlates with the Boolean complexity of the concept. This suggests that in-context learning exhibits a learning bias for simplicity in a way similar to humans.