paper-with-me

Papers

CoMP: Continual Multimodal Pre-training for Vision Foundation Models

2025-03-24 · Yitong Chen, Lingchen Meng, Wujian Peng, Zuxuan Wu, Yu-Gang Jiang

Pre-trained Vision Foundation Models (VFMs) provide strong visual representations for a wide range of applications. In this paper, we continually pre-train prevailing VFMs in a multimodal manner such that they can effortlessly process visual inputs of varying sizes and produce visual representations that are more aligned with language representations, regardless of their original pre-training process. To this end, we introduce CoMP, a carefully designed multimodal pre-training pipeline. CoMP uses a Continual Rotary Position Embedding to accommodate visual inputs with different resolutions, and an Alignment Loss between visual and textual features for better cross-modal alignment. After continual pre-training, leading VFMs like DINOv2, SigLIP and AIMv2 achieve remarkable improvements not only in multimodal understanding tasks but also in generic classification and segmentation tasks. Remarkably, CoMP-AIMv2 achieves scores of 64.9 on ChartQA with a 0.5B LLM, while maintaining an 87.3% accuracy on ImageNet-1K and a 51.8 mIoU on ADE20K under frozen chunk evaluation.

📄 PDF Abstract BibTeX arXiv:2503.18931

Code (1)

SliMM-X/CoMP-MM 공식 구현 pytorch

Tasks

cross-modal alignment

Similar Papers 제목 키워드 기반

A Practitioner's Guide to Continual Multimodal Pretraining

2024-08-26 · Karsten Roth, Vishaal Udandarao, Sebastian Dziadzio, Ameya Prabhu 외

Multimodal foundation models serve numerous applications at the intersection of vision and language. Still, despite being pretrained on extensive data, they become outdated over time. To keep models updated, research int…

Continual LearningContinual PretrainingMeta-Learning

FTibSuite: A Comprehensive Resource Suite for Tibetan Vision-Language Modeling

2026-05-26 · Guixian Xu, Yide Liang, Zeli Su, Xuexian Song 외 arxiv

Vision-language models have progressed rapidly, but Tibetan remains a severely underserved low-resource language due to the lack of reproducible training and evaluation infrastructure. To fill this gap, we introduce FTib…

Continual Pretraining

Towards Robust and Fair Vision Learning in Open-World Environments

2024-12-12 · Thanh-Dat Truong

The dissertation presents four key contributions toward fairness and robustness in vision learning. First, to address the problem of large-scale data requirements, the dissertation presents a novel Fairness Domain Adapta…

Continual LearningDomain AdaptationDomain GeneralizationFairness

Bisecle: Binding and Separation in Continual Learning for Video Language Understanding

2025-07-01 · Yue Tan, Xiaoqian Hu, Hao Xue, Celso De Melo 외 arxiv

Frontier vision-language models (VLMs) have made remarkable improvements in video understanding tasks. However, real-world videos typically exist as continuously evolving data streams (e.g., dynamic scenes captured by we…

Continual Learning

Parameter Importance-Driven Continual Learning for Foundation Models

2025-11-19 · Lingxiang Wang, Hainan Zhang, Zhiming Zheng arxiv

Domain-specific post-training often causes catastrophic forgetting, making foundation models lose their general reasoning ability and limiting their adaptability to dynamic real-world environments. Preserving general cap…

Continual Learning