Language Model Crossover: Variation through Few-Shot Prompting

TL;DR

This paper introduces language model crossover, a novel method leveraging few-shot prompting to evolve text-based representations, demonstrating versatility across various data types and promising for semantic evolution.

Contribution

It presents a simple, open-source compatible variation operator using language models for evolving text-based genomes, expanding evolutionary techniques into NLP and code domains.

Findings

Effective in evolving binary strings, sentences, and code

Versatile across multiple data types and representations

Leverages current language model advancements

Abstract

This paper pursues the insight that language models naturally enable an intelligent variation operator similar in spirit to evolutionary crossover. In particular, language models of sufficient scale demonstrate in-context learning, i.e. they can learn from associations between a small number of input patterns to generate outputs incorporating such associations (also called few-shot prompting). This ability can be leveraged to form a simple but powerful variation operator, i.e. to prompt a language model with a few text-based genotypes (such as code, plain-text sentences, or equations), and to parse its corresponding output as those genotypes' offspring. The promise of such language model crossover (which is simple to implement and can leverage many different open-source language models) is that it enables a simple mechanism to evolve semantically-rich text representations (with few domain-specific tweaks), and naturally benefits from current progress in language models. Experiments in this paper highlight the versatility of language-model crossover, through evolving binary bit-strings, sentences, equations, text-to-image prompts, and Python code. The conclusion is that language model crossover is a promising method for evolving genomes representable as text.