Longhui Wei

papers

1,204

total citations

papers (17)

Visformer: The Vision-Friendly Transformer

ICCV 2021arXiv

277

citations

Rethinking the Distribution Gap of Person Re-identification with Camera-based Batch Normalization

ECCV 2020arXiv

181

citations

Gradient-Regulated Meta-Prompt Learning for Generalizable Vision-Language Models

ICCV 2023arXiv

citations

Integrally Pre-Trained Transformer Pyramid Networks

CVPR 2023arXiv

citations

Network Adjustment: Channel Search Guided by FLOPs Utilization Ratio

CVPR 2020arXiv

citations

Boosting Segment Anything Model Towards Open-Vocabulary Learning

AAAI 2025arXiv

citations

C-CLIP: Multimodal Continual Learning for Vision-Language Model

ICLR 2025

citations

Efficient Multi-modal Long Context Learning for Training-free Adaptation

ICML 2025arXiv

citations

OVMR: Open-Vocabulary Recognition with Multi-Modal References

CVPR 2024arXiv

citations

Learning To Learn by Jointly Optimizing Neural Architecture and Weights

CVPR 2022

citations

Longhui Wei

papers (17)

Visformer: The Vision-Friendly Transformer

Rethinking the Distribution Gap of Person Re-identification with Camera-based Batch Normalization

HalluciDoctor: Mitigating Hallucinatory Toxicity in Visual Instruction Data

MVP: Multimodality-Guided Visual Pre-training

Rectifying the Shortcut Learning of Background for Few-Shot Learning

Fine-Grained Semantically Aligned Vision-Language Pre-Training

UnrealPerson: An Adaptive Pipeline Towards Costless Person Re-Identification

Circumventing Outliers of AutoAugment with Knowledge Distillation

Enhance Image Classification via Inter-Class Image Mixup with Diffusion Model

Gradient-Regulated Meta-Prompt Learning for Generalizable Vision-Language Models

Integrally Pre-Trained Transformer Pyramid Networks

Network Adjustment: Channel Search Guided by FLOPs Utilization Ratio

Boosting Segment Anything Model Towards Open-Vocabulary Learning

C-CLIP: Multimodal Continual Learning for Vision-Language Model

Efficient Multi-modal Long Context Learning for Training-free Adaptation

OVMR: Open-Vocabulary Recognition with Multi-Modal References

Learning To Learn by Jointly Optimizing Neural Architecture and Weights

papers (17)

Visformer: The Vision-Friendly Transformer

Rethinking the Distribution Gap of Person Re-identification with Camera-based Batch Normalization

HalluciDoctor: Mitigating Hallucinatory Toxicity in Visual Instruction Data

MVP: Multimodality-Guided Visual Pre-training

Rectifying the Shortcut Learning of Background for Few-Shot Learning

Fine-Grained Semantically Aligned Vision-Language Pre-Training

UnrealPerson: An Adaptive Pipeline Towards Costless Person Re-Identification

Circumventing Outliers of AutoAugment with Knowledge Distillation

Enhance Image Classification via Inter-Class Image Mixup with Diffusion Model

Gradient-Regulated Meta-Prompt Learning for Generalizable Vision-Language Models

Integrally Pre-Trained Transformer Pyramid Networks

Network Adjustment: Channel Search Guided by FLOPs Utilization Ratio

Boosting Segment Anything Model Towards Open-Vocabulary Learning

C-CLIP: Multimodal Continual Learning for Vision-Language Model

Efficient Multi-modal Long Context Learning for Training-free Adaptation

OVMR: Open-Vocabulary Recognition with Multi-Modal References

Learning To Learn by Jointly Optimizing Neural Architecture and Weights