Improving Structural Diversity of Blackbox LLMs via Chain-of-Specification Prompting

TL;DR

This paper introduces a new way to measure and improve the diversity of large language models' outputs by allowing user-specified features and using a chain-of-specification prompting strategy, effective even with blackbox models.

Contribution

The paper proposes a novel structural diversity metric and a chain-of-specification prompting method to enhance diversity control in blackbox LLMs.

Findings

CoS improves structural diversity in poetry and code generation.

The new diversity metric aligns with user-defined features.

CoS outperforms baseline methods in diversity metrics.

Abstract

The capability to generate diverse text is a key challenge facing large language models (LLMs). Thus far, diversity has been studied via metrics such as $n$-gram diversity or diversity of BERT embeddings. However, for these kinds of diversity, the user has little control over the dimensions along which diversity is considered. For example, in the poetry domain, one might desire diversity in terms of rhyme and meter, whereas in the code domain, one might desire diversity in terms of the kinds of expressions used to solve a problem. We propose a diversity metric called structural diversity, where the user provides a mapping from generated text to features capturing the kinds of diversity that they care about. In addition, we propose a novel strategy called chain-of-specification (CoS) prompting for improving diversity by first having the LLM generate a specification encoding one instance of structural features, and then prompting the LLM to generate text that satisfies these features; notably, our strategy works with blackbox LLMs. In our experiments, we show that for structural diversity in the poetry and code domains, CoS significantly improves diversity compared to several baselines.