paper-with-me

Papers

TIPSv2: Advancing Vision-Language Pretraining with Enhanced Patch-Text Alignment

2026-04-13 · Bingyi Cao, Koert Chen, Kevis-Kokitsi Maninis, Kaifeng Chen, Arjun Karpur, Ye Xia, Sahil Dua, Tanmaya Dabral, Guangxing Han, Bohyung Han, Joshua Ainslie, Alex Bewley, Mithun Jacob, René Wagner, Washington Ramos, Krzysztof Choromanski, Mojtaba Seyedhosseini, Howard Zhou, André Araujo arxiv

Recent progress in vision-language pretraining has enabled significant improvements to many downstream computer vision applications, such as classification, retrieval, segmentation and depth prediction. However, a fundamental capability that these models still struggle with is aligning dense patch representations with text embeddings of corresponding concepts. In this work, we investigate this critical issue and propose novel techniques to enhance this capability in foundational vision-language models. First, we reveal that a patch-level distillation procedure significantly boosts dense patch-text alignment -- surprisingly, the patch-text alignment of the distilled student model strongly surpasses that of the teacher model. This observation inspires us to consider modifications to pretraining recipes, leading us to propose iBOT++, an upgrade to the commonly-used iBOT masked image objective, where unmasked tokens also contribute directly to the loss. This dramatically enhances patch-text alignment of pretrained models. Additionally, to improve vision-language pretraining efficiency and effectiveness, we modify the exponential moving average setup in the learning recipe, and introduce a caption sampling strategy to benefit from synthetic captions at different granularities. Combining these components, we develop TIPSv2, a new family of image-text encoder models suitable for a wide range of downstream applications. Through comprehensive experiments on 9 tasks and 20 datasets, we demonstrate strong performance, generally on par with or better than recent vision encoder models. Code and models are released via our project page at https://gdm-tipsv2.github.io/ .

📄 PDF Abstract BibTeX arXiv:2604.12012

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

MAKE: Multi-Aspect Knowledge-Enhanced Vision-Language Pretraining for Zero-shot Dermatological Assessment

2025-05-14 · Siyuan Yan, Xieji Li, Ming Hu, Yiwen Jiang 외

Dermatological diagnosis represents a complex multimodal challenge that requires integrating visual features with specialized clinical knowledge. While vision-language pretraining (VLP) has advanced medical AI, its effec…

Clinical KnowledgeContrastive LearningCross-Modal Retrieval

Improving Medical Visual Representations via Radiology Report Generation

2023-10-30 · Keegan Quigley, Miriam Cha, Josh Barua, Geeticka Chauhan 외

Vision-language pretraining has been shown to produce high-quality visual encoders which transfer efficiently to downstream computer vision tasks. Contrastive learning approaches have increasingly been adopted for medica…

Contrastive LearningDecoderImage CaptioningMedical Image Analysis

Beyond Language Modeling: An Exploration of Multimodal Pretraining

2026-03-03 · Shengbang Tong, David Fan, John Nguyen, Ellis Brown 외 arxiv

The visual world offers a critical axis for advancing foundation models beyond language. Despite growing interest in this direction, the design space for native multimodal models remains opaque. We provide empirical clar…

M2-Encoder: Advancing Bilingual Image-Text Understanding by Large-scale Efficient Pretraining

2024-01-29 · Qingpei Guo, Furong Xu, Hanxiao Zhang, Wang Ren 외

Vision-language foundation models like CLIP have revolutionized the field of artificial intelligence. Nevertheless, VLM models supporting multi-language, e.g., in both Chinese and English, have lagged due to the relative…

GPUzero-shot-classificationZero-Shot Cross-Modal RetrievalZero-shot Image Retrieval+3

Supervised Fine-tuning in turn Improves Visual Foundation Models

2024-01-18 · Xiaohu Jiang, Yixiao Ge, Yuying Ge, Dachuan Shi 외

Image-text training like CLIP has dominated the pretraining of vision foundation models in recent years. Subsequent efforts have been made to introduce region-level visual learning into CLIP's pretraining but face scalab…