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

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Yadav, Vikas, Tang, Zheng, Srinivasan, Vijay
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866911930986266624
author Yadav, Vikas
Tang, Zheng
Srinivasan, Vijay
author_facet Yadav, Vikas
Tang, Zheng
Srinivasan, Vijay
contents 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
format Preprint
id arxiv_https___arxiv_org_abs_2406_17163
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Paraphrase and Aggregate with Large Language Models for Minimizing Intent Classification Errors
Yadav, Vikas
Tang, Zheng
Srinivasan, Vijay
Computation and Language
Artificial Intelligence
Machine Learning
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
title Paraphrase and Aggregate with Large Language Models for Minimizing Intent Classification Errors
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2406.17163