[CLS] is Not Enough: Multi-Label Recognition via Patch-Level Inference and Adaptive Aggregation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Akang, Deng, Xili, Hu, Zhanxuan, Zhao, Yi, Tai, Yonghang, Li, Huafeng
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910255191949312
author Wang, Akang
Deng, Xili
Hu, Zhanxuan
Zhao, Yi
Tai, Yonghang
Li, Huafeng
author_facet Wang, Akang
Deng, Xili
Hu, Zhanxuan
Zhao, Yi
Tai, Yonghang
Li, Huafeng
contents Vision-Language Models such as CLIP exhibit strong zero-shot recognition capability by aligning images with textual concepts, yet they often underperform on multi-label recognition where multiple objects co-exist. A key bottleneck is that the [CLS] token, as a single global visual representation, is insufficient to faithfully encode diverse targets with varying scales, contexts, and co-occurrence patterns. To address this limitation, we present a new multi-label image recognition framework, termed PIAA, which formulates prediction as Patch-level Inference followed by Adaptive Aggregation. Specifically, we first enhance patch-wise predictions from two complementary perspectives: (i) mitigating semantic entanglement in the visual encoder to obtain more discriminative patch representations, and (ii) learning an unsupervised visual classifier to narrow the vision-language modality gap. We then introduce an adaptive aggregation module that consolidates patch-level scores into the final multi-label prediction. Notably, the entire pipeline is fully training-free, requiring no gradient updates or parameter fine-tuning. Experiments show that our method achieves strong improvements with minimal extra computation, exceeding a 6% mAP gain on the challenging NUS-WIDE benchmark over representative baselines. Code is available at https://github.com/akang-wang/PIAA.
format Preprint
id arxiv_https___arxiv_org_abs_2605_25821
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle [CLS] is Not Enough: Multi-Label Recognition via Patch-Level Inference and Adaptive Aggregation
Wang, Akang
Deng, Xili
Hu, Zhanxuan
Zhao, Yi
Tai, Yonghang
Li, Huafeng
Computer Vision and Pattern Recognition
Vision-Language Models such as CLIP exhibit strong zero-shot recognition capability by aligning images with textual concepts, yet they often underperform on multi-label recognition where multiple objects co-exist. A key bottleneck is that the [CLS] token, as a single global visual representation, is insufficient to faithfully encode diverse targets with varying scales, contexts, and co-occurrence patterns. To address this limitation, we present a new multi-label image recognition framework, termed PIAA, which formulates prediction as Patch-level Inference followed by Adaptive Aggregation. Specifically, we first enhance patch-wise predictions from two complementary perspectives: (i) mitigating semantic entanglement in the visual encoder to obtain more discriminative patch representations, and (ii) learning an unsupervised visual classifier to narrow the vision-language modality gap. We then introduce an adaptive aggregation module that consolidates patch-level scores into the final multi-label prediction. Notably, the entire pipeline is fully training-free, requiring no gradient updates or parameter fine-tuning. Experiments show that our method achieves strong improvements with minimal extra computation, exceeding a 6% mAP gain on the challenging NUS-WIDE benchmark over representative baselines. Code is available at https://github.com/akang-wang/PIAA.
title [CLS] is Not Enough: Multi-Label Recognition via Patch-Level Inference and Adaptive Aggregation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2605.25821