# TagCLIP: A Local-to-Global Framework to Enhance Open-Vocabulary Multi-Label Classification of CLIP Without Training

arXiv: https://arxiv.org/abs/2312.12828 · HTML: https://mlsanity.com/paper/2312.12828/

## Key facts

- Cited by 19 papers in the map (data through 14 Oct 2026).
- Builds on 43 of its 48 references that are in the map.
- First posted on arXiv 20 Dec 2023.

## Builds on (most cited first, 30 of 43)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks](https://mlsanity.com/paper/1506.01497/) (2015)
- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs](https://mlsanity.com/paper/1606.00915/) (2016)
- [Learning Deep Features for Discriminative Localization](https://mlsanity.com/paper/1512.04150/) (2015)
- [Self-training with Noisy Student improves ImageNet classification](https://mlsanity.com/paper/1911.04252/) (2019)
- [Open-vocabulary Object Detection via Vision and Language Knowledge Distillation](https://mlsanity.com/paper/2104.13921/) (2021)
- [COCO-Stuff: Thing and Stuff Classes in Context](https://mlsanity.com/paper/1612.03716/) (2016)
- [Do Vision Transformers See Like Convolutional Neural Networks?](https://mlsanity.com/paper/2108.08810/) (2021)
- [RegionCLIP: Region-based Language-Image Pretraining](https://mlsanity.com/paper/2112.09106/) (2021)
- [Extract Free Dense Labels from CLIP](https://mlsanity.com/paper/2112.01071/) (2021)
- [GroupViT: Semantic Segmentation Emerges from Text Supervision](https://mlsanity.com/paper/2202.11094/) (2022)
- [Invariant Information Clustering for Unsupervised Image Classification and Segmentation](https://mlsanity.com/paper/1807.06653/) (2018)
- [Asymmetric Loss For Multi-Label Classification](https://mlsanity.com/paper/2009.14119/) (2020)
- [Rethinking Pre-training and Self-training](https://mlsanity.com/paper/2006.06882/) (2020)
- [Multi-Label Image Recognition with Graph Convolutional Networks](https://mlsanity.com/paper/1904.03582/) (2019)
- [Learning Pixel-level Semantic Affinity with Image-level Supervision for Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/1803.10464/) (2018)
- [Self-supervised Equivariant Attention Mechanism for Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/2004.04581/) (2020)
- [VQGAN-CLIP: Open Domain Image Generation and Editing with Natural Language Guidance](https://mlsanity.com/paper/2204.08583/) (2022)
- [Unsupervised Semantic Segmentation by Contrasting Object Mask Proposals](https://mlsanity.com/paper/2102.06191/) (2021)
- [SegCLIP: Patch Aggregation with Learnable Centers for Open-Vocabulary Semantic Segmentation](https://mlsanity.com/paper/2211.14813/) (2022)
- [PiCIE: Unsupervised Semantic Segmentation using Invariance and Equivariance in Clustering](https://mlsanity.com/paper/2103.17070/) (2021)
- [TS-CAM: Token Semantic Coupled Attention Map for Weakly Supervised Object Localization](https://mlsanity.com/paper/2103.14862/) (2021)
- [Distribution-Balanced Loss for Multi-Label Classification in Long-Tailed Datasets](https://mlsanity.com/paper/2007.09654/) (2020)
- [Learning Affinity from Attention: End-to-End Weakly-Supervised Semantic Segmentation with Transformers](https://mlsanity.com/paper/2203.02664/) (2022)
- [ReCo: Retrieve and Co-segment for Zero-shot Transfer](https://mlsanity.com/paper/2206.07045/) (2022)
- [Learning Semantic-Specific Graph Representation for Multi-Label Image Recognition](https://mlsanity.com/paper/1908.07325/) (2019)
- [DualCoOp: Fast Adaptation to Multi-Label Recognition with Limited Annotations](https://mlsanity.com/paper/2206.09541/) (2022)
- [SegSort: Segmentation by Discriminative Sorting of Segments](https://mlsanity.com/paper/1910.06962/) (2019)

## Built on by (most cited first, 19 of 19)

- [Self-Calibrated CLIP for Training-Free Open-Vocabulary Segmentation](https://mlsanity.com/paper/2411.15869/) (2024)
- [CLIPer: Hierarchically Improving Spatial Representation of CLIP for Open-Vocabulary Semantic Segmentation](https://mlsanity.com/paper/2411.13836/) (2024)
- [iSeg: An Iterative Refinement-based Framework for Training-free Segmentation](https://mlsanity.com/paper/2409.03209/) (2024)
- [A Review of Multimodal Explainable Artificial Intelligence: Past, Present and Future](https://mlsanity.com/paper/2412.14056/) (2024)
- [ITACLIP: Boosting Training-Free Semantic Segmentation with Image, Text, and Architectural Enhancements](https://mlsanity.com/paper/2411.12044/) (2024)
- [Not Just What's There: Enabling CLIP to Comprehend Negated Visual Descriptions Without Fine-tuning](https://mlsanity.com/paper/2602.21035/) (2026)
- [Are Multimodal Large Language Models Good Annotators for Image Tagging?](https://mlsanity.com/paper/2602.20972/) (2026)
- [Query-Based Knowledge Sharing for Open-Vocabulary Multi-Label Classification](https://mlsanity.com/paper/2401.01181/) (2024)
- [Test-time Contrastive Concepts for Open-world Semantic Segmentation with Vision-Language Models](https://mlsanity.com/paper/2407.05061/) (2024)
- [Exploring Causes and Mitigation of Hallucinations in Large Vision Language Models](https://mlsanity.com/paper/2502.16842/) (2025)
- [SPARC: Score Prompting and Adaptive Fusion for Zero-Shot Multi-Label Recognition in Vision-Language Models](https://mlsanity.com/paper/2502.16911/) (2025)
- [NERVE: Neighbourhood & Entropy-guided Random-walk for training free open-Vocabulary sEgmentation](https://mlsanity.com/paper/2511.08248/) (2025)
- [Category-Adaptive Cross-Modal Semantic Refinement and Transfer for Open-Vocabulary Multi-Label Recognition](https://mlsanity.com/paper/2412.06190/) (2024)
- [DART: Dual Adaptive Refinement Transfer for Open-Vocabulary Multi-Label Recognition](https://mlsanity.com/paper/2508.05585/) (2025)
- [AeroLite: Tag-Guided Lightweight Generation of Aerial Image Captions](https://mlsanity.com/paper/2504.09528/) (2025)
- [microCLIP: Unsupervised CLIP Adaptation via Coarse-Fine Token Fusion for Fine-Grained Image Classification](https://mlsanity.com/paper/2510.02270/) (2025)
- [[CLS] is Not Enough: Multi-Label Recognition via Patch-Level Inference and Adaptive Aggregation](https://mlsanity.com/paper/2605.25821/) (2026)
- [RA-ClipScore: Making Generative Model Evaluation More Interpretable](https://mlsanity.com/paper/2608.12088/) (2026)
- [Towards Purified Multi-Label Test-Time Adaptation of Vision-Language Models](https://mlsanity.com/paper/2608.25653/) (2026)

## Abstract (arXiv)

Contrastive Language-Image Pre-training (CLIP) has demonstrated impressive capabilities in open-vocabulary classification. The class token in the image encoder is trained to capture the global features to distinguish different text descriptions supervised by contrastive loss, making it highly effective for single-label classification. However, it shows poor performance on multi-label datasets because the global feature tends to be dominated by the most prominent class and the contrastive nature of softmax operation aggravates it. In this study, we observe that the multi-label classification results heavily rely on discriminative local features but are overlooked by CLIP. As a result, we dissect the preservation of patch-wise spatial information in CLIP and proposed a local-to-global framework to obtain image tags. It comprises three steps: (1) patch-level classification to obtain coarse scores; (2) dual-masking attention refinement (DMAR) module to refine the coarse scores; (3) class-wise reidentification (CWR) module to remedy predictions from a global perspective. This framework is solely based on frozen CLIP and significantly enhances its multi-label classification performance on various benchmarks without dataset-specific training. Besides, to comprehensively assess the quality and practicality of generated tags, we extend their application to the downstream task, i.e., weakly supervised semantic segmentation (WSSS) with generated tags as image-level pseudo labels. Experiments demonstrate that this classify-then-segment paradigm dramatically outperforms other annotation-free segmentation methods and validates the effectiveness of generated tags. Our code is available at https://github.com/linyq2117/TagCLIP.