Rethinking Interpretability in the Era of Large Language Models

TL;DR

This paper discusses how large language models can transform interpretability in machine learning by enabling natural language explanations, while addressing new challenges and proposing future research directions.

Contribution

It reviews current evaluation methods for LLM interpretability and advocates for leveraging LLMs to analyze datasets and generate interactive explanations.

Findings

LLMs can explain in natural language, expanding interpretability scope.

Existing evaluation methods have limitations but are foundational.

Future research should focus on dataset analysis and interactive explanations.

Abstract

Interpretable machine learning has exploded as an area of interest over the last decade, sparked by the rise of increasingly large datasets and deep neural networks. Simultaneously, large language models (LLMs) have demonstrated remarkable capabilities across a wide array of tasks, offering a chance to rethink opportunities in interpretable machine learning. Notably, the capability to explain in natural language allows LLMs to expand the scale and complexity of patterns that can be given to a human. However, these new capabilities raise new challenges, such as hallucinated explanations and immense computational costs. In this position paper, we start by reviewing existing methods to evaluate the emerging field of LLM interpretation (both interpreting LLMs and using LLMs for explanation). We contend that, despite their limitations, LLMs hold the opportunity to redefine interpretability with a more ambitious scope across many applications, including in auditing LLMs themselves. We highlight two emerging research priorities for LLM interpretation: using LLMs to directly analyze new datasets and to generate interactive explanations.