Saved in:
Bibliographic Details
Main Authors: Vakharia, Priyesh, Joshi, Devavrat, Chavan, Meenal, Sonawane, Dhananjay, Garg, Bhrigu, Mazaheri, Parsa
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2312.14346
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917629248143360
author Vakharia, Priyesh
Joshi, Devavrat
Chavan, Meenal
Sonawane, Dhananjay
Garg, Bhrigu
Mazaheri, Parsa
author_facet Vakharia, Priyesh
Joshi, Devavrat
Chavan, Meenal
Sonawane, Dhananjay
Garg, Bhrigu
Mazaheri, Parsa
contents Large Language Models (LLMs) are adept at text manipulation -- tasks such as machine translation and text summarization. However, these models can also be prone to hallucination, which can be detrimental to the faithfulness of any answers that the model provides. Recent works in combating hallucinations in LLMs deal with identifying hallucinated sentences and categorizing the different ways in which models hallucinate. This paper takes a deep dive into LLM behavior with respect to hallucinations, defines a token-level approach to identifying different kinds of hallucinations, and further utilizes this token-level tagging to improve the interpretability and faithfulness of LLMs in dialogue summarization tasks. Through this, the paper presents a new, enhanced dataset and a new training paradigm.
format Preprint
id arxiv_https___arxiv_org_abs_2312_14346
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Don't Believe Everything You Read: Enhancing Summarization Interpretability through Automatic Identification of Hallucinations in Large Language Models
Vakharia, Priyesh
Joshi, Devavrat
Chavan, Meenal
Sonawane, Dhananjay
Garg, Bhrigu
Mazaheri, Parsa
Computation and Language
Artificial Intelligence
Large Language Models (LLMs) are adept at text manipulation -- tasks such as machine translation and text summarization. However, these models can also be prone to hallucination, which can be detrimental to the faithfulness of any answers that the model provides. Recent works in combating hallucinations in LLMs deal with identifying hallucinated sentences and categorizing the different ways in which models hallucinate. This paper takes a deep dive into LLM behavior with respect to hallucinations, defines a token-level approach to identifying different kinds of hallucinations, and further utilizes this token-level tagging to improve the interpretability and faithfulness of LLMs in dialogue summarization tasks. Through this, the paper presents a new, enhanced dataset and a new training paradigm.
title Don't Believe Everything You Read: Enhancing Summarization Interpretability through Automatic Identification of Hallucinations in Large Language Models
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2312.14346