Bohan Zhuang

papers

1,462

total citations

papers (27)

EfficientDM: Efficient Quantization-Aware Fine-Tuning of Low-Bit Diffusion Models

ICLR 2024arXiv

citations

Training Quantized Neural Networks With a Full-Precision Auxiliary Module

CVPR 2020arXiv

citations

Automated Progressive Learning for Efficient Training of Vision Transformers

CVPR 2022arXiv

citations

BiViT: Extremely Compressed Binary Vision Transformers

ICCV 2023arXiv

citations

Stitchable Neural Networks

CVPR 2023arXiv

citations

EcoFormer: Energy-Saving Attention with Linear Complexity

NEURIPS 2022arXiv

citations

An Efficient Spatio-Temporal Pyramid Transformer for Action Detection

ECCV 2022arXiv

citations

Efficient Test-Time Adaptation for Super-Resolution with Second-Order Degradation and Reconstruction

NEURIPS 2023arXiv

citations

Dynamic Focus-Aware Positional Queries for Semantic Segmentation

CVPR 2023arXiv

citations

Mask Propagation for Efficient Video Semantic Segmentation

NEURIPS 2023arXiv

citations

ModaVerse: Efficiently Transforming Modalities with LLMs

CVPR 2024arXiv

citations

FATNN: Fast and Accurate Ternary Neural Networks

ICCV 2021arXiv

citations

Neighboring Autoregressive Modeling for Efficient Visual Generation

ICCV 2025arXiv

citations

AQD: Towards Accurate Quantized Object Detection

CVPR 2021arXiv

citations

FPSAttention: Training-Aware FP8 and Sparsity Co-Design for Fast Video Diffusion

NEURIPS 2025arXiv

citations

Efficient Stitchable Task Adaptation

CVPR 2024arXiv

citations

ZPressor: Bottleneck-Aware Compression for Scalable Feed-Forward 3DGS

NEURIPS 2025arXiv

citations

Frequency-Aware Autoregressive Modeling for Efficient High-Resolution Image Synthesis

ICCV 2025arXiv

citations

Stitched ViTs are Flexible Vision Backbones

ECCV 2024arXiv

citations

Channel Merging: Preserving Specialization for Merged Experts

AAAI 2025arXiv

citations

ZipVL: Accelerating Vision-Language Models through Dynamic Token Sparsity

ICCV 2025

citations

Bohan Zhuang

papers (27)

Fast Vision Transformers with HiLo Attention

PTQD: Accurate Post-Training Quantization for Diffusion Models

Scalable Vision Transformers With Hierarchical Pooling

Generative Low-bitwidth Data Free Quantization

LongVLM: Efficient Long Video Understanding via Large Language Models

Sensitivity-Aware Visual Parameter-Efficient Fine-Tuning

EfficientDM: Efficient Quantization-Aware Fine-Tuning of Low-Bit Diffusion Models

Training Quantized Neural Networks With a Full-Precision Auxiliary Module

Automated Progressive Learning for Efficient Training of Vision Transformers

BiViT: Extremely Compressed Binary Vision Transformers

Stitchable Neural Networks

EcoFormer: Energy-Saving Attention with Linear Complexity

An Efficient Spatio-Temporal Pyramid Transformer for Action Detection

Efficient Test-Time Adaptation for Super-Resolution with Second-Order Degradation and Reconstruction

Dynamic Focus-Aware Positional Queries for Semantic Segmentation

Mask Propagation for Efficient Video Semantic Segmentation

ModaVerse: Efficiently Transforming Modalities with LLMs

FATNN: Fast and Accurate Ternary Neural Networks

Neighboring Autoregressive Modeling for Efficient Visual Generation

AQD: Towards Accurate Quantized Object Detection

FPSAttention: Training-Aware FP8 and Sparsity Co-Design for Fast Video Diffusion

Efficient Stitchable Task Adaptation

ZPressor: Bottleneck-Aware Compression for Scalable Feed-Forward 3DGS

Frequency-Aware Autoregressive Modeling for Efficient High-Resolution Image Synthesis

Stitched ViTs are Flexible Vision Backbones

Channel Merging: Preserving Specialization for Merged Experts

ZipVL: Accelerating Vision-Language Models through Dynamic Token Sparsity

papers (27)

Fast Vision Transformers with HiLo Attention

PTQD: Accurate Post-Training Quantization for Diffusion Models

Scalable Vision Transformers With Hierarchical Pooling

Generative Low-bitwidth Data Free Quantization

LongVLM: Efficient Long Video Understanding via Large Language Models

Sensitivity-Aware Visual Parameter-Efficient Fine-Tuning

EfficientDM: Efficient Quantization-Aware Fine-Tuning of Low-Bit Diffusion Models

Training Quantized Neural Networks With a Full-Precision Auxiliary Module

Automated Progressive Learning for Efficient Training of Vision Transformers

BiViT: Extremely Compressed Binary Vision Transformers

Stitchable Neural Networks

EcoFormer: Energy-Saving Attention with Linear Complexity

An Efficient Spatio-Temporal Pyramid Transformer for Action Detection

Efficient Test-Time Adaptation for Super-Resolution with Second-Order Degradation and Reconstruction

Dynamic Focus-Aware Positional Queries for Semantic Segmentation

Mask Propagation for Efficient Video Semantic Segmentation

ModaVerse: Efficiently Transforming Modalities with LLMs

FATNN: Fast and Accurate Ternary Neural Networks

Neighboring Autoregressive Modeling for Efficient Visual Generation

AQD: Towards Accurate Quantized Object Detection

FPSAttention: Training-Aware FP8 and Sparsity Co-Design for Fast Video Diffusion

Efficient Stitchable Task Adaptation

ZPressor: Bottleneck-Aware Compression for Scalable Feed-Forward 3DGS

Frequency-Aware Autoregressive Modeling for Efficient High-Resolution Image Synthesis

Stitched ViTs are Flexible Vision Backbones

Channel Merging: Preserving Specialization for Merged Experts

ZipVL: Accelerating Vision-Language Models through Dynamic Token Sparsity