SegSub: Evaluating Robustness to Knowledge Conflicts and Hallucinations in Vision-Language Models

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Carragher, Peter, Rao, Nikitha, Jha, Abhinand, Raghav, R, Carley, Kathleen M.
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866912486906658816
author Carragher, Peter
Rao, Nikitha
Jha, Abhinand
Raghav, R
Carley, Kathleen M.
author_facet Carragher, Peter
Rao, Nikitha
Jha, Abhinand
Raghav, R
Carley, Kathleen M.
contents Vision language models (VLM) demonstrate sophisticated multimodal reasoning yet are prone to hallucination when confronted with knowledge conflicts, impeding their deployment in information-sensitive contexts. While existing research addresses robustness in unimodal models, the multimodal domain lacks systematic investigation of cross-modal knowledge conflicts. This research introduces \segsub, a framework for applying targeted image perturbations to investigate VLM resilience against knowledge conflicts. Our analysis reveals distinct vulnerability patterns: while VLMs are robust to parametric conflicts (20% adherence rates), they exhibit significant weaknesses in identifying counterfactual conditions (<30% accuracy) and resolving source conflicts (<1% accuracy). Correlations between contextual richness and hallucination rate (r = -0.368, p = 0.003) reveal the kinds of images that are likely to cause hallucinations. Through targeted fine-tuning on our benchmark dataset, we demonstrate improvements in VLM knowledge conflict detection, establishing a foundation for developing hallucination-resilient multimodal systems in information-sensitive environments.
format Preprint
id arxiv_https___arxiv_org_abs_2502_14908
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SegSub: Evaluating Robustness to Knowledge Conflicts and Hallucinations in Vision-Language Models
Carragher, Peter
Rao, Nikitha
Jha, Abhinand
Raghav, R
Carley, Kathleen M.
Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
Machine Learning
Vision language models (VLM) demonstrate sophisticated multimodal reasoning yet are prone to hallucination when confronted with knowledge conflicts, impeding their deployment in information-sensitive contexts. While existing research addresses robustness in unimodal models, the multimodal domain lacks systematic investigation of cross-modal knowledge conflicts. This research introduces \segsub, a framework for applying targeted image perturbations to investigate VLM resilience against knowledge conflicts. Our analysis reveals distinct vulnerability patterns: while VLMs are robust to parametric conflicts (20% adherence rates), they exhibit significant weaknesses in identifying counterfactual conditions (<30% accuracy) and resolving source conflicts (<1% accuracy). Correlations between contextual richness and hallucination rate (r = -0.368, p = 0.003) reveal the kinds of images that are likely to cause hallucinations. Through targeted fine-tuning on our benchmark dataset, we demonstrate improvements in VLM knowledge conflict detection, establishing a foundation for developing hallucination-resilient multimodal systems in information-sensitive environments.
title SegSub: Evaluating Robustness to Knowledge Conflicts and Hallucinations in Vision-Language Models
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
Machine Learning
url https://arxiv.org/abs/2502.14908