Large Language Models Are Latent Variable Models: Explaining and Finding Good Demonstrations for In-Context Learning

TL;DR

This paper models large language models as latent variable models to explain in-context learning and proposes an algorithm to select effective demonstrations, significantly improving performance across multiple datasets and models.

Contribution

It introduces a Bayesian perspective to understand LLMs as latent variable models and presents a demonstration selection algorithm that enhances in-context learning.

Findings

Improved performance over baselines on eight GPT models and datasets

Effective demonstration selection enhances in-context learning

Supports the hypothesis that LLMs infer latent task variables

Abstract

In recent years, pre-trained large language models (LLMs) have demonstrated remarkable efficiency in achieving an inference-time few-shot learning capability known as in-context learning. However, existing literature has highlighted the sensitivity of this capability to the selection of few-shot demonstrations. Current understandings of the underlying mechanisms by which this capability arises from regular language model pretraining objectives remain disconnected from the real-world LLMs. This study aims to examine the in-context learning phenomenon through a Bayesian lens, viewing real-world LLMs as latent variable models. On this premise, we propose an algorithm to select optimal demonstrations from a set of annotated data with a small LM, and then directly generalize the selected demonstrations to larger LMs. We demonstrate significant improvement over baselines, averaged over eight GPT models on eight real-world text classification datasets. We also demonstrate the real-world usefulness of our algorithm on GSM8K, a math word problem dataset. Our empirical findings support our hypothesis that LLMs implicitly infer a latent variable containing task information.