@inproceedings{cao-etal-2024-visdiahalbench,
title = "{V}is{D}ia{H}al{B}ench: A Visual Dialogue Benchmark For Diagnosing Hallucination in Large Vision-Language Models",
author = "Cao, Qingxing and
Cheng, Junhao and
Liang, Xiaodan and
Lin, Liang",
editor = "Ku, Lun-Wei and
Martins, Andre and
Srikumar, Vivek",
booktitle = "Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = aug,
year = "2024",
address = "Bangkok, Thailand",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.acl-long.658",
pages = "12161--12176",
abstract = "Despite the significant success of large vision-language models (LVLMs), some studies have revealed that LVLMs suffer from the hallucination problem, where the LVLMs{'} response contains descriptions of non-existent objects. Although various benchmarks have been proposed to investigate this problem, they mostly focus on single-turn evaluation and overlook the hallucination raised by textual inputs. To investigate the hallucination problem of LVLMs when given long-term misleading textual history, we propose a novel visual dialogue hallucination evaluation benchmark VisDiaHalBench. The benchmark consists of samples with five-turn questions about an edited image and its original version. VisDiaHalBench differs from previous hallucination benchmarks in the following three points: 1) The questions and answers are unambiguously grounded by annotated scene graphs. 2) The images are uncommonly edited to inspect the visual model and common-object hallucination in LLMs. 3) The carefully designed dialogue refers a same object in different turns to assess the image consistency and influence of history for LVLMs. The detailed analysis of several state-of-the-art LVLMs across image consistency, visual understanding, history influence, and other dimensions reveals their substantial performance gap with single-turn VQA tasks. The benchmark is released in: https://github.com/qingxingcao/VisDiaHalBench",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="cao-etal-2024-visdiahalbench">
<titleInfo>
<title>VisDiaHalBench: A Visual Dialogue Benchmark For Diagnosing Hallucination in Large Vision-Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Qingxing</namePart>
<namePart type="family">Cao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Junhao</namePart>
<namePart type="family">Cheng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaodan</namePart>
<namePart type="family">Liang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Liang</namePart>
<namePart type="family">Lin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lun-Wei</namePart>
<namePart type="family">Ku</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andre</namePart>
<namePart type="family">Martins</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vivek</namePart>
<namePart type="family">Srikumar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Bangkok, Thailand</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Despite the significant success of large vision-language models (LVLMs), some studies have revealed that LVLMs suffer from the hallucination problem, where the LVLMs’ response contains descriptions of non-existent objects. Although various benchmarks have been proposed to investigate this problem, they mostly focus on single-turn evaluation and overlook the hallucination raised by textual inputs. To investigate the hallucination problem of LVLMs when given long-term misleading textual history, we propose a novel visual dialogue hallucination evaluation benchmark VisDiaHalBench. The benchmark consists of samples with five-turn questions about an edited image and its original version. VisDiaHalBench differs from previous hallucination benchmarks in the following three points: 1) The questions and answers are unambiguously grounded by annotated scene graphs. 2) The images are uncommonly edited to inspect the visual model and common-object hallucination in LLMs. 3) The carefully designed dialogue refers a same object in different turns to assess the image consistency and influence of history for LVLMs. The detailed analysis of several state-of-the-art LVLMs across image consistency, visual understanding, history influence, and other dimensions reveals their substantial performance gap with single-turn VQA tasks. The benchmark is released in: https://github.com/qingxingcao/VisDiaHalBench</abstract>
<identifier type="citekey">cao-etal-2024-visdiahalbench</identifier>
<location>
<url>https://aclanthology.org/2024.acl-long.658</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>12161</start>
<end>12176</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T VisDiaHalBench: A Visual Dialogue Benchmark For Diagnosing Hallucination in Large Vision-Language Models
%A Cao, Qingxing
%A Cheng, Junhao
%A Liang, Xiaodan
%A Lin, Liang
%Y Ku, Lun-Wei
%Y Martins, Andre
%Y Srikumar, Vivek
%S Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand
%F cao-etal-2024-visdiahalbench
%X Despite the significant success of large vision-language models (LVLMs), some studies have revealed that LVLMs suffer from the hallucination problem, where the LVLMs’ response contains descriptions of non-existent objects. Although various benchmarks have been proposed to investigate this problem, they mostly focus on single-turn evaluation and overlook the hallucination raised by textual inputs. To investigate the hallucination problem of LVLMs when given long-term misleading textual history, we propose a novel visual dialogue hallucination evaluation benchmark VisDiaHalBench. The benchmark consists of samples with five-turn questions about an edited image and its original version. VisDiaHalBench differs from previous hallucination benchmarks in the following three points: 1) The questions and answers are unambiguously grounded by annotated scene graphs. 2) The images are uncommonly edited to inspect the visual model and common-object hallucination in LLMs. 3) The carefully designed dialogue refers a same object in different turns to assess the image consistency and influence of history for LVLMs. The detailed analysis of several state-of-the-art LVLMs across image consistency, visual understanding, history influence, and other dimensions reveals their substantial performance gap with single-turn VQA tasks. The benchmark is released in: https://github.com/qingxingcao/VisDiaHalBench
%U https://aclanthology.org/2024.acl-long.658
%P 12161-12176
Markdown (Informal)
[VisDiaHalBench: A Visual Dialogue Benchmark For Diagnosing Hallucination in Large Vision-Language Models](https://aclanthology.org/2024.acl-long.658) (Cao et al., ACL 2024)
ACL