analiza aiqbench
This research introduces UHP Detection, a novel black-box framework that identifies hallucinations in Large Vision-Language Models by analyzing structured uncertainty patterns across image and text perturbations rather than relying on single consistency metrics. By leveraging four complementary consistency groups derived from logical polarity and modality variations, the method significantly outperforms existing baselines in detecting ungrounded visual predictions without requiring model access. This advancement offers a more robust tool for validating multimodal reasoning systems where hallucination detection is critical for deployment reliability.
UHP Detection: LVLMs have their Unique Hallucination Pattern in the Consistency Space
arXiv:2608.03817v1 Announce Type: cross Abstract: Large vision--language models (LVLMs) demonstrate strong multimodal reasoning capabilities but remain prone to hallucination, where model predictions are not grounded in visual evidence. Existing black-box hallucination detection methods estimate uncertainty through a single consistency metric, implicitly assuming that model uncertainty can be adequately characterized by a single measure. However, hallucinations exhibit diverse manifestations of uncertainty across different behavioral probes, making a single measure insufficient to characterize their underlying behavior. We propose \emph{Unique Hallucination Pattern (UHP) Detection}, a fully black-box framework that models hallucination as a structured uncertainty pattern defined by two axes: perturbation modality (image vs.\ text) and logical polarity (a statement vs.\ its negation). Their intersection produces four complementary consistency groups that capture distinct manifestations of model uncertainty, from which both within-group and between-group features are extracted to train a lightweight classifier. Through comprehensive experiments on AMBER and PhD across three LVLMs, UHP Detection consistently outperforms prior black-box and white-box baselines, with improvements of up to $+18.72\%$ AUC-ROC and $+20.07\%$ AUC-PR over the strongest black-box methods. Extensive ablation studies demonstrate that each consistency group contributes complementary information and that their combination forms a structured hallucination pattern. Furthermore, cross-dataset evaluation shows that this learned pattern generalizes across benchmarks, indicating that hallucination behavior reflects a model-specific consistency pattern. \textbf{Code is publicly available at} https://github.com/amirezzati/uhpdet.
Abstract:Large vision--language models (LVLMs) demonstrate strong multimodal reasoning capabilities but remain prone to hallucination, where model predictions are not grounded in visual evidence. Existing black-box hallucination detection methods estimate uncertainty through a single consistency metric, implicitly assuming that model uncertainty can be adequately characterized by a single measure. However, hallucinations exhibit diverse manifestations of uncertainty across different behavioral probes, making a single measure insufficient to characterize their underlying behavior. We propose \emph{Unique Hallucination Pattern (UHP) Detection}, a fully black-box framework that models hallucination as a structured uncertainty pattern defined by two axes: perturbation modality (image vs.\ text) and logical polarity (a statement vs.\ its negation). Their intersection produces four complementary consistency groups that capture distinct manifestations of model uncertainty, from which both within-group and between-group features are extracted to train a lightweight classifier. Through comprehensive experiments on AMBER and PhD across three LVLMs, UHP Detection consistently outperforms prior black-box and white-box baselines, with improvements of up to $+18.72\%$ AUC-ROC and $+20.07\%$ AUC-PR over the strongest black-box methods. Extensive ablation studies demonstrate that each consistency group contributes complementary information and that their combination forms a structured hallucination pattern. Furthermore, cross-dataset evaluation shows that this learned pattern generalizes across benchmarks, indicating that hallucination behavior reflects a model-specific consistency pattern. \textbf{Code is publicly available at} this https URL.
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
arXiv:2608.03817 [cs.CV]
(or arXiv:2608.03817v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2608.03817 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 315 cuvinte
Subiecte conexe
AgentOps: Operationalize agentic AI at scale with Amazon Bedrock AgentCore
AWS · 1 iun. 2026
What We Learned Trying to Catch AI Liars: An Aletheia's Quest Retrospective
EleutherAI · 25 aug. 2026
SpanCalib-VLM: Calibrated Hallucination Span Detection in Vision-Language Models
arXiv · 1 sept. 2026
Construct validity of Claude Opus 4.8's System Card – A commentary
LessWrong · 12 iun. 2026