LLM Processes: Numerical Predictive Distributions Conditioned on Natural Language

TL;DR

This paper introduces LLM Processes, a method for generating probabilistic numerical predictions conditioned on natural language, enabling more nuanced, expert-informed regression models leveraging large language models.

Contribution

The paper presents a novel approach to eliciting coherent predictive distributions from LLMs conditioned on natural language, expanding their use in probabilistic regression and related tasks.

Findings

Effective elicitation of predictive distributions from LLMs

Improved regression performance with natural language guidance

Demonstrated applicability across forecasting, optimization, and image modeling

Abstract

Machine learning practitioners often face significant challenges in formally integrating their prior knowledge and beliefs into predictive models, limiting the potential for nuanced and context-aware analyses. Moreover, the expertise needed to integrate this prior knowledge into probabilistic modeling typically limits the application of these models to specialists. Our goal is to build a regression model that can process numerical data and make probabilistic predictions at arbitrary locations, guided by natural language text which describes a user's prior knowledge. Large Language Models (LLMs) provide a useful starting point for designing such a tool since they 1) provide an interface where users can incorporate expert insights in natural language and 2) provide an opportunity for leveraging latent problem-relevant knowledge encoded in LLMs that users may not have themselves. We start by exploring strategies for eliciting explicit, coherent numerical predictive distributions from LLMs. We examine these joint predictive distributions, which we call LLM Processes, over arbitrarily-many quantities in settings such as forecasting, multi-dimensional regression, black-box optimization, and image modeling. We investigate the practical details of prompting to elicit coherent predictive distributions, and demonstrate their effectiveness at regression. Finally, we demonstrate the ability to usefully incorporate text into numerical predictions, improving predictive performance and giving quantitative structure that reflects qualitative descriptions. This lets us begin to explore the rich, grounded hypothesis space that LLMs implicitly encode.