13 papers
SAB-LVLM: Significance-Aware Binarization for Large Vision-Language Models
Qi Lyu, Jiahua Dong, Baichen Liu +7
Large Vision-Language Models (LVLMs) have achieved remarkable progress in multimodal understanding, yet their enormous parameter scale and cross-modal computation incur substantial…
Does YOLO Really Need to See Every Training Image in Every Epoch?
Xingxing Xie, Jiahua Dong, Junwei Han +1
YOLO detectors are known for their fast inference speed, yet training them remains unexpectedly time-consuming due to their exhaustive pipeline that processes every training image…
Bring Your Dreams to Life: Continual Text-to-Video Customization
Jiahua Dong, Xudong Wang, Wenqi Liang +7
Customized text-to-video generation (CTVG) has recently witnessed great progress in generating tailored videos from user-specific text. However, most CTVG methods assume that perso…
Thinking with Drafts: Speculative Temporal Reasoning for Efficient Long Video Understanding
Pengfei Hu, Meng Cao, Yingyao Wang +6
Long video understanding is essential for human-like intelligence, enabling coherent perception and reasoning over extended temporal contexts. While the emerging thinking-with-fram…
CRISP: Contrastive Residual Injection and Semantic Prompting for Continual Video Instance Segmentation
Baichen Liu, Qi Lyu, Xudong Wang +3
Continual video instance segmentation demands both the plasticity to absorb new object categories and the stability to retain previously learned ones, all while preserving temporal…
Hierarchical Visual Prompt Learning for Continual Video Instance Segmentation
Jiahua Dong, Hui Yin, Wenqi Liang +5
Video instance segmentation (VIS) has gained significant attention for its capability in tracking and segmenting object instances across video frames. However, most of the existing…
IAP: Improving Continual Learning of Vision-Language Models via Instance-Aware Prompting
Hao Fu, Hanbin Zhao, Jiahua Dong +3
Recent pre-trained vision-language models (PT-VLMs) often face a Multi-Domain Task Incremental Learning (MTIL) scenario in practice, where several classes and domains of multi-moda…
Video SimpleQA: Towards Factuality Evaluation in Large Video Language Models
Meng Cao, Pengfei Hu, Yingyao Wang +12
Recent advancements in Large Video Language Models (LVLMs) have highlighted their potential for multi-modal understanding, yet evaluating their factual grounding in videos remains…
CE-SDWV: Effective and Efficient Concept Erasure for Text-to-Image Diffusion Models via a Semantic-Driven Word Vocabulary
Jiahang Tu, Qian Feng, Jiahua Dong +4
Large-scale text-to-image (T2I) diffusion models have achieved remarkable generative performance about various concepts. With the limitation of privacy and safety in practice, the…
Continual LLaVA: Continual Instruction Tuning in Large Vision-Language Models
Meng Cao, Yuyang Liu, Yingfei Liu +6
Instruction tuning constitutes a prevalent technique for tailoring Large Vision Language Models (LVLMs) to meet individual task requirements. To date, most of the existing approach…
How to Continually Adapt Text-to-Image Diffusion Models for Flexible Customization?
Jiahua Dong, Wenqi Liang, Hongliu Li +5
Custom diffusion models (CDMs) have attracted widespread attention due to their astonishing generative ability for personalized concepts. However, most existing CDMs unreasonably a…
PECTP: Parameter-Efficient Cross-Task Prompts for Incremental Vision Transformer
Qian Feng, Hanbin Zhao, Chao Zhang +4
Incremental Learning (IL) aims to learn deep models on sequential tasks continually, where each new task includes a batch of new classes and deep models have no access to task-ID i…
Dual Hyperspectral Mamba for Efficient Spectral Compressive Imaging
Jiahua Dong, Hui Yin, Hongliu Li +4
Deep unfolding methods have made impressive progress in restoring 3D hyperspectral images (HSIs) from 2D measurements through convolution neural networks or Transformers in spectra…