FactCHD: Benchmarking Fact-Conflicting Hallucination Detection

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Chen, Xiang, Song, Duanzheng, Gui, Honghao, Wang, Chenxi, Zhang, Ningyu, Jiang, Yong, Huang, Fei, Lv, Chengfei, Zhang, Dan, Chen, Huajun
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914812324216832
author Chen, Xiang
Song, Duanzheng
Gui, Honghao
Wang, Chenxi
Zhang, Ningyu
Jiang, Yong
Huang, Fei
Lv, Chengfei
Zhang, Dan
Chen, Huajun
author_facet Chen, Xiang
Song, Duanzheng
Gui, Honghao
Wang, Chenxi
Zhang, Ningyu
Jiang, Yong
Huang, Fei
Lv, Chengfei
Zhang, Dan
Chen, Huajun
contents Despite their impressive generative capabilities, LLMs are hindered by fact-conflicting hallucinations in real-world applications. The accurate identification of hallucinations in texts generated by LLMs, especially in complex inferential scenarios, is a relatively unexplored area. To address this gap, we present FactCHD, a dedicated benchmark designed for the detection of fact-conflicting hallucinations from LLMs. FactCHD features a diverse dataset that spans various factuality patterns, including vanilla, multi-hop, comparison, and set operation. A distinctive element of FactCHD is its integration of fact-based evidence chains, significantly enhancing the depth of evaluating the detectors' explanations. Experiments on different LLMs expose the shortcomings of current approaches in detecting factual errors accurately. Furthermore, we introduce Truth-Triangulator that synthesizes reflective considerations by tool-enhanced ChatGPT and LoRA-tuning based on Llama2, aiming to yield more credible detection through the amalgamation of predictive results and evidence. The benchmark dataset is available at https://github.com/zjunlp/FactCHD.
format Preprint
id arxiv_https___arxiv_org_abs_2310_12086
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle FactCHD: Benchmarking Fact-Conflicting Hallucination Detection
Chen, Xiang
Song, Duanzheng
Gui, Honghao
Wang, Chenxi
Zhang, Ningyu
Jiang, Yong
Huang, Fei
Lv, Chengfei
Zhang, Dan
Chen, Huajun
Computation and Language
Artificial Intelligence
Computer Vision and Pattern Recognition
Information Retrieval
Machine Learning
Despite their impressive generative capabilities, LLMs are hindered by fact-conflicting hallucinations in real-world applications. The accurate identification of hallucinations in texts generated by LLMs, especially in complex inferential scenarios, is a relatively unexplored area. To address this gap, we present FactCHD, a dedicated benchmark designed for the detection of fact-conflicting hallucinations from LLMs. FactCHD features a diverse dataset that spans various factuality patterns, including vanilla, multi-hop, comparison, and set operation. A distinctive element of FactCHD is its integration of fact-based evidence chains, significantly enhancing the depth of evaluating the detectors' explanations. Experiments on different LLMs expose the shortcomings of current approaches in detecting factual errors accurately. Furthermore, we introduce Truth-Triangulator that synthesizes reflective considerations by tool-enhanced ChatGPT and LoRA-tuning based on Llama2, aiming to yield more credible detection through the amalgamation of predictive results and evidence. The benchmark dataset is available at https://github.com/zjunlp/FactCHD.
title FactCHD: Benchmarking Fact-Conflicting Hallucination Detection
topic Computation and Language
Artificial Intelligence
Computer Vision and Pattern Recognition
Information Retrieval
Machine Learning
url https://arxiv.org/abs/2310.12086