Research

Paper

TESTING March 03, 2026

Mind the Way You Select Negative Texts: Pursuing the Distance Consistency in OOD Detection with VLMs

Authors

Zhikang Xu, Qianqian Xu, Zitai Wang, Cong Hua, Sicong Li, Zhiyong Yang, Qingming Huang

Abstract

Out-of-distribution (OOD) detection seeks to identify samples from unknown classes, a critical capability for deploying machine learning models in open-world scenarios. Recent research has demonstrated that Vision-Language Models (VLMs) can effectively leverage their multi-modal representations for OOD detection. However, current methods often incorporate intra-modal distance during OOD detection, such as comparing negative texts with ID labels or comparing test images with image proxies. This design paradigm creates an inherent inconsistency against the inter-modal distance that CLIP-like VLMs are optimized for, potentially leading to suboptimal performance. To address this limitation, we propose InterNeg, a simple yet effective framework that systematically utilizes consistent inter-modal distance enhancement from textual and visual perspectives. From the textual perspective, we devise an inter-modal criterion for selecting negative texts. From the visual perspective, we dynamically identify high-confidence OOD images and invert them into the textual space, generating extra negative text embeddings guided by inter-modal distance. Extensive experiments across multiple benchmarks demonstrate the superiority of our approach. Notably, our InterNeg achieves state-of-the-art performance compared to existing works, with a 3.47\% reduction in FPR95 on the large-scale ImageNet benchmark and a 5.50\% improvement in AUROC on the challenging Near-OOD benchmark.

Metadata

arXiv ID: 2603.02618
Provider: ARXIV
Primary Category: cs.CV
Published: 2026-03-03
Fetched: 2026-03-04 03:41

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.02618v1</id>\n    <title>Mind the Way You Select Negative Texts: Pursuing the Distance Consistency in OOD Detection with VLMs</title>\n    <updated>2026-03-03T05:44:47Z</updated>\n    <link href='https://arxiv.org/abs/2603.02618v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.02618v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Out-of-distribution (OOD) detection seeks to identify samples from unknown classes, a critical capability for deploying machine learning models in open-world scenarios. Recent research has demonstrated that Vision-Language Models (VLMs) can effectively leverage their multi-modal representations for OOD detection. However, current methods often incorporate intra-modal distance during OOD detection, such as comparing negative texts with ID labels or comparing test images with image proxies. This design paradigm creates an inherent inconsistency against the inter-modal distance that CLIP-like VLMs are optimized for, potentially leading to suboptimal performance. To address this limitation, we propose InterNeg, a simple yet effective framework that systematically utilizes consistent inter-modal distance enhancement from textual and visual perspectives. From the textual perspective, we devise an inter-modal criterion for selecting negative texts. From the visual perspective, we dynamically identify high-confidence OOD images and invert them into the textual space, generating extra negative text embeddings guided by inter-modal distance. Extensive experiments across multiple benchmarks demonstrate the superiority of our approach. Notably, our InterNeg achieves state-of-the-art performance compared to existing works, with a 3.47\\% reduction in FPR95 on the large-scale ImageNet benchmark and a 5.50\\% improvement in AUROC on the challenging Near-OOD benchmark.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n    <published>2026-03-03T05:44:47Z</published>\n    <arxiv:comment>Accepted by the main track of CVPR 2026</arxiv:comment>\n    <arxiv:primary_category term='cs.CV'/>\n    <author>\n      <name>Zhikang Xu</name>\n    </author>\n    <author>\n      <name>Qianqian Xu</name>\n    </author>\n    <author>\n      <name>Zitai Wang</name>\n    </author>\n    <author>\n      <name>Cong Hua</name>\n    </author>\n    <author>\n      <name>Sicong Li</name>\n    </author>\n    <author>\n      <name>Zhiyong Yang</name>\n    </author>\n    <author>\n      <name>Qingming Huang</name>\n    </author>\n  </entry>"
}