Cover Image

Continual Learning with Vision-Language Models via Semantic-Geometry Preservation

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	He, Chiyuan, Qiu, Zihuan, Meng, Fanman, Zhang, Runtong, Xu, Linfeng, Wu, Qingbo, Li, Hongliang
Format:	Preprint
Published:	2026
Subjects:	Computer Vision and Pattern Recognition Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

MINGLE: Mixture of Null-Space Gated Low-Rank Experts for Test-Time Continual Model Merging
by: Qiu, Zihuan, et al.
Published: (2025)

DesCLIP: Robust Continual Learning via General Attribute Descriptions for VLM-Based Visual Recognition
by: He, Chiyuan, et al.
Published: (2025)

Closing the Oracle Gap: Increment Vector Transformation for Class Incremental Learning
by: Qiu, Zihuan, et al.
Published: (2025)

FireRescue: A UAV-Based Dataset and Enhanced YOLO Model for Object Detection in Fire Rescue Scenes
by: Xu, Qingyu, et al.
Published: (2025)

Towards Continual Egocentric Activity Recognition: A Multi-modal Egocentric Activity Dataset for Continual Learning
by: Xu, Linfeng, et al.
Published: (2023)

No Re-Train, More Gain: Upgrading Backbones with Diffusion model for Pixel-Wise and Weakly-Supervised Few-Shot Segmentation
by: Chen, Shuai, et al.
Published: (2024)

Distribution-Level Memory Recall for Continual Learning: Preserving Knowledge and Avoiding Confusion
by: Cheng, Shaoxu, et al.
Published: (2024)

Few-Shot Continual Learning for Activity Recognition in Classroom Surveillance Images
by: Qian, Yilei, et al.
Published: (2024)

Cognition Transferring and Decoupling for Text-supervised Egocentric Semantic Segmentation
by: Shi, Zhaofeng, et al.
Published: (2024)

Null-Space Filtering for Data-Free Continual Model Merging: Preserving Stability, Promoting Plasticity
by: Qiu, Zihuan, et al.
Published: (2025)

Leveraging Pre-Trained Models for Multimodal Class-Incremental Learning under Adaptive Fusion
by: Chen, Yukun, et al.
Published: (2025)

CMaP-SAM: Contraction Mapping Prior for SAM-driven Few-shot Segmentation
by: Chen, Shuai, et al.
Published: (2025)

Test-time Ego-Exo-centric Adaptation for Action Anticipation via Multi-Label Prototype Growing and Dual-Clue Consistency
by: Shi, Zhaofeng, et al.
Published: (2026)

SAVA-X: Ego-to-Exo Imitation Error Detection via Scene-Adaptive View Alignment and Bidirectional Cross View Fusion
by: Li, Xiang, et al.
Published: (2026)

Attention-disentangled Uniform Orthogonal Feature Space Optimization for Few-shot Object Detection
by: Zhao, Taijin, et al.
Published: (2025)

CMP: A Composable Meta Prompt for SAM-Based Cross-Domain Few-Shot Segmentation
by: Chen, Shuai, et al.
Published: (2025)

DFR: A Decompose-Fuse-Reconstruct Framework for Multi-Modal Few-Shot Segmentation
by: Chen, Shuai, et al.
Published: (2025)

SS-DC: Spatial-Spectral Decoupling and Coupling Across Visible-Infrared Gap for Domain Adaptive Object Detection
by: Zhang, Xiwei, et al.
Published: (2025)

Quantifying Cross-Modality Memorization in Vision-Language Models
by: Wen, Yuxin, et al.
Published: (2025)

Privacy-Preserving Synthetic Continual Semantic Segmentation for Robotic Surgery
by: Xu, Mengya, et al.
Published: (2024)

GRSDet: Learning to Generate Local Reverse Samples for Few-shot Object Detection
by: Mei, Hefei, et al.
Published: (2023)

WD-FQDet: Multispectral Detection Transformer via Wavelet Decomposition and Frequency-aware Query Learning
by: Yang, Chunjin, et al.
Published: (2026)

Continual Learning in Vision-Language Models via Aligned Model Merging
by: Sokar, Ghada, et al.
Published: (2025)

An Experimental Study of Semantic Continuity for Deep Learning Models
by: Wu, Shangxi, et al.
Published: (2020)

Synthetic Data is an Elegant GIFT for Continual Vision-Language Models
by: Wu, Bin, et al.
Published: (2025)

Learning Self-Correction in Vision-Language Models via Rollout Augmentation
by: Ding, Yi, et al.
Published: (2026)

ARIC: An Activity Recognition Dataset in Classroom Surveillance Images
by: Xu, Linfeng, et al.
Published: (2024)

Multi-Stage Knowledge Integration of Vision-Language Models for Continual Learning
by: Zhang, Hongsheng, et al.
Published: (2024)

Mind the Gap: Preserving and Compensating for the Modality Gap in CLIP-Based Continual Learning
by: Huang, Linlan, et al.
Published: (2025)

Continuity-Preserving Convolutional Autoencoders for Learning Continuous Latent Dynamical Models from Images
by: Zhu, Aiqing, et al.
Published: (2025)

MASC: Boosting Autoregressive Image Generation with a Manifold-Aligned Semantic Clustering
by: He, Lixuan, et al.
Published: (2025)

Let Language Constrain Geometry: Vision-Language Models as Semantic and Spatial Critics for 3D Generation
by: Bai, Weimin, et al.
Published: (2025)

Cross-Modal Redundancy and the Geometry of Vision-Language Embeddings
by: Dhimoïla, Grégoire, et al.
Published: (2026)

Semantic Document Derendering: SVG Reconstruction via Vision-Language Modeling
by: Hazimeh, Adam, et al.
Published: (2025)

Beyond Perception Errors: Semantic Fixation in Large Vision-Language Models
by: Alam, Md Tanvirul
Published: (2026)

Boosting Continual Learning of Vision-Language Models via Mixture-of-Experts Adapters
by: Yu, Jiazuo, et al.
Published: (2024)

VERA: Explainable Video Anomaly Detection via Verbalized Learning of Vision-Language Models
by: Ye, Muchao, et al.
Published: (2024)

Does Visual Information Play a Decisive Role in Vision-Language-Action Model Driving Behavior?
by: He, Jingtao, et al.
Published: (2026)

Semantic Compositions Enhance Vision-Language Contrastive Learning
by: Aladago, Maxwell, et al.
Published: (2024)

Learning to Rank Pre-trained Vision-Language Models for Downstream Tasks
by: Ding, Yuhe, et al.
Published: (2024)