Paraphrase and Aggregate with Large Language Models for Minimizing Intent Classification Errors

TL;DR

This paper introduces PAG-LLM, a method that uses paraphrasing and aggregation with large language models to improve intent classification accuracy and reduce errors, especially for difficult cases.

Contribution

The paper proposes a novel paraphrase-and-aggregate approach with LLMs to enhance classification accuracy and minimize errors in multi-class intent classification tasks.

Findings

Achieved 22.7% and 15.1% error reduction on two datasets.

Effective in reducing misclassification and hallucinated labels.

Improves performance on hard-to-classify examples.

Abstract

Large language models (LLM) have achieved remarkable success in natural language generation but lesser focus has been given to their applicability in decision making tasks such as classification. We show that LLMs like LLaMa can achieve high performance on large multi-class classification tasks but still make classification errors and worse, generate out-of-vocabulary class labels. To address these critical issues, we introduce Paraphrase and AGgregate (PAG)-LLM approach wherein an LLM generates multiple paraphrases of the input query (parallel queries), performs multi-class classification for the original query and each paraphrase, and at the end aggregate all the classification labels based on their confidence scores. We evaluate PAG-LLM on two large multi-class classication datasets: CLINC, and Banking and show 22.7% and 15.1% error reduction. We show that PAG-LLM is especially effective for hard examples where LLM is uncertain, and reduces the critical misclassification and hallucinated label generation errors