ConVis: Contrastive Decoding with Hallucination Visualization for Mitigating Hallucinations in Multimodal Large Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Park, Yeji, Lee, Deokyeong, Choe, Junsuk, Chang, Buru
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913480319172608
author Park, Yeji
Lee, Deokyeong
Choe, Junsuk
Chang, Buru
author_facet Park, Yeji
Lee, Deokyeong
Choe, Junsuk
Chang, Buru
contents Hallucinations in Multimodal Large Language Models (MLLMs) where generated responses fail to accurately reflect the given image pose a significant challenge to their reliability. To address this, we introduce ConVis, a novel training-free contrastive decoding method. ConVis leverages a text-to-image (T2I) generation model to semantically reconstruct the given image from hallucinated captions. By comparing the contrasting probability distributions produced by the original and reconstructed images, ConVis enables MLLMs to capture visual contrastive signals that penalize hallucination generation. Notably, this method operates purely within the decoding process, eliminating the need for additional data or model updates. Our extensive experiments on five popular benchmarks demonstrate that ConVis effectively reduces hallucinations across various MLLMs, highlighting its potential to enhance model reliability.
format Preprint
id arxiv_https___arxiv_org_abs_2408_13906
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle ConVis: Contrastive Decoding with Hallucination Visualization for Mitigating Hallucinations in Multimodal Large Language Models
Park, Yeji
Lee, Deokyeong
Choe, Junsuk
Chang, Buru
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Hallucinations in Multimodal Large Language Models (MLLMs) where generated responses fail to accurately reflect the given image pose a significant challenge to their reliability. To address this, we introduce ConVis, a novel training-free contrastive decoding method. ConVis leverages a text-to-image (T2I) generation model to semantically reconstruct the given image from hallucinated captions. By comparing the contrasting probability distributions produced by the original and reconstructed images, ConVis enables MLLMs to capture visual contrastive signals that penalize hallucination generation. Notably, this method operates purely within the decoding process, eliminating the need for additional data or model updates. Our extensive experiments on five popular benchmarks demonstrate that ConVis effectively reduces hallucinations across various MLLMs, highlighting its potential to enhance model reliability.
title ConVis: Contrastive Decoding with Hallucination Visualization for Mitigating Hallucinations in Multimodal Large Language Models
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2408.13906