Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Chen, Lixian, Chen, Yanhui, Lin, Junyi
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2604.24602
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915975516913664
author Chen, Lixian
Chen, Yanhui
Lin, Junyi
author_facet Chen, Lixian
Chen, Yanhui
Lin, Junyi
contents Vision-language models transfer well in zero-shot settings, but at deployment the visual and textual branches often shift asymmetrically. Under this condition, entropy-based test-time adaptation can sharpen the fused posterior while increasing error, because an unreliable modality may still dominate fusion. We study this failure mode through a majorization view of multimodal posteriors and cast adaptation as a constrained de-mixing problem on the fused prediction. Based on this view, we propose MG-MTTA, which keeps the backbone frozen and updates only a lightweight gate or adapter. The objective combines fused-posterior entropy minimization with a reliability-aware gate prior built from anchor-based modality consistency and cross-modal conflict. Our analysis gives conditions under which entropy reduction preserves the correct ranking and a threshold that characterizes modality-dominance failure. On the ImageNet-based benchmark, MG-MTTA improves top-1 accuracy from 57.97 to 66.51 under semantics-preserving textual shift and from 21.68 to 26.27 under joint visual-textual shift, while remaining competitive in the visual-only benchmark. These results show that multimodal test-time adaptation should control modality reliability, not just prediction entropy.
format Preprint
id arxiv_https___arxiv_org_abs_2604_24602
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Majorization-Guided Test-Time Adaptation for Vision-Language Models under Modality-Specific Shift
Chen, Lixian
Chen, Yanhui
Lin, Junyi
Computer Vision and Pattern Recognition
Vision-language models transfer well in zero-shot settings, but at deployment the visual and textual branches often shift asymmetrically. Under this condition, entropy-based test-time adaptation can sharpen the fused posterior while increasing error, because an unreliable modality may still dominate fusion. We study this failure mode through a majorization view of multimodal posteriors and cast adaptation as a constrained de-mixing problem on the fused prediction. Based on this view, we propose MG-MTTA, which keeps the backbone frozen and updates only a lightweight gate or adapter. The objective combines fused-posterior entropy minimization with a reliability-aware gate prior built from anchor-based modality consistency and cross-modal conflict. Our analysis gives conditions under which entropy reduction preserves the correct ranking and a threshold that characterizes modality-dominance failure. On the ImageNet-based benchmark, MG-MTTA improves top-1 accuracy from 57.97 to 66.51 under semantics-preserving textual shift and from 21.68 to 26.27 under joint visual-textual shift, while remaining competitive in the visual-only benchmark. These results show that multimodal test-time adaptation should control modality reliability, not just prediction entropy.
title Majorization-Guided Test-Time Adaptation for Vision-Language Models under Modality-Specific Shift
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2604.24602