Alert button
Picture for Kaisheng Ma

Kaisheng Ma

Alert button

DreamLLM: Synergistic Multimodal Comprehension and Creation

Sep 20, 2023
Runpei Dong, Chunrui Han, Yuang Peng, Zekun Qi, Zheng Ge, Jinrong Yang, Liang Zhao, Jianjian Sun, Hongyu Zhou, Haoran Wei, Xiangwen Kong, Xiangyu Zhang, Kaisheng Ma, Li Yi

Figure 1 for DreamLLM: Synergistic Multimodal Comprehension and Creation
Figure 2 for DreamLLM: Synergistic Multimodal Comprehension and Creation
Figure 3 for DreamLLM: Synergistic Multimodal Comprehension and Creation
Figure 4 for DreamLLM: Synergistic Multimodal Comprehension and Creation
Viaarxiv icon

VPP: Efficient Conditional 3D Generation via Voxel-Point Progressive Representation

Jul 28, 2023
Zekun Qi, Muzhou Yu, Runpei Dong, Kaisheng Ma

Figure 1 for VPP: Efficient Conditional 3D Generation via Voxel-Point Progressive Representation
Figure 2 for VPP: Efficient Conditional 3D Generation via Voxel-Point Progressive Representation
Figure 3 for VPP: Efficient Conditional 3D Generation via Voxel-Point Progressive Representation
Figure 4 for VPP: Efficient Conditional 3D Generation via Voxel-Point Progressive Representation
Viaarxiv icon

Point-GCC: Universal Self-supervised 3D Scene Pre-training via Geometry-Color Contrast

Jun 01, 2023
Guofan Fan, Zekun Qi, Wenkai Shi, Kaisheng Ma

Figure 1 for Point-GCC: Universal Self-supervised 3D Scene Pre-training via Geometry-Color Contrast
Figure 2 for Point-GCC: Universal Self-supervised 3D Scene Pre-training via Geometry-Color Contrast
Figure 3 for Point-GCC: Universal Self-supervised 3D Scene Pre-training via Geometry-Color Contrast
Figure 4 for Point-GCC: Universal Self-supervised 3D Scene Pre-training via Geometry-Color Contrast
Viaarxiv icon

Revisiting Data Augmentation in Model Compression: An Empirical and Comprehensive Study

May 22, 2023
Muzhou Yu, Linfeng Zhang, Kaisheng Ma

Figure 1 for Revisiting Data Augmentation in Model Compression: An Empirical and Comprehensive Study
Figure 2 for Revisiting Data Augmentation in Model Compression: An Empirical and Comprehensive Study
Figure 3 for Revisiting Data Augmentation in Model Compression: An Empirical and Comprehensive Study
Figure 4 for Revisiting Data Augmentation in Model Compression: An Empirical and Comprehensive Study
Viaarxiv icon

CORSD: Class-Oriented Relational Self Distillation

Apr 28, 2023
Muzhou Yu, Sia Huat Tan, Kailu Wu, Runpei Dong, Linfeng Zhang, Kaisheng Ma

Figure 1 for CORSD: Class-Oriented Relational Self Distillation
Figure 2 for CORSD: Class-Oriented Relational Self Distillation
Figure 3 for CORSD: Class-Oriented Relational Self Distillation
Figure 4 for CORSD: Class-Oriented Relational Self Distillation
Viaarxiv icon

CLIP-FO3D: Learning Free Open-world 3D Scene Representations from 2D Dense CLIP

Mar 16, 2023
Junbo Zhang, Runpei Dong, Kaisheng Ma

Figure 1 for CLIP-FO3D: Learning Free Open-world 3D Scene Representations from 2D Dense CLIP
Figure 2 for CLIP-FO3D: Learning Free Open-world 3D Scene Representations from 2D Dense CLIP
Figure 3 for CLIP-FO3D: Learning Free Open-world 3D Scene Representations from 2D Dense CLIP
Figure 4 for CLIP-FO3D: Learning Free Open-world 3D Scene Representations from 2D Dense CLIP
Viaarxiv icon

Hebbian and Gradient-based Plasticity Enables Robust Memory and Rapid Learning in RNNs

Feb 07, 2023
Yu Duan, Zhongfan Jia, Qian Li, Yi Zhong, Kaisheng Ma

Figure 1 for Hebbian and Gradient-based Plasticity Enables Robust Memory and Rapid Learning in RNNs
Figure 2 for Hebbian and Gradient-based Plasticity Enables Robust Memory and Rapid Learning in RNNs
Figure 3 for Hebbian and Gradient-based Plasticity Enables Robust Memory and Rapid Learning in RNNs
Figure 4 for Hebbian and Gradient-based Plasticity Enables Robust Memory and Rapid Learning in RNNs
Viaarxiv icon

Contrast with Reconstruct: Contrastive 3D Representation Learning Guided by Generative Pretraining

Feb 05, 2023
Zekun Qi, Runpei Dong, Guofan Fan, Zheng Ge, Xiangyu Zhang, Kaisheng Ma, Li Yi

Figure 1 for Contrast with Reconstruct: Contrastive 3D Representation Learning Guided by Generative Pretraining
Figure 2 for Contrast with Reconstruct: Contrastive 3D Representation Learning Guided by Generative Pretraining
Figure 3 for Contrast with Reconstruct: Contrastive 3D Representation Learning Guided by Generative Pretraining
Figure 4 for Contrast with Reconstruct: Contrastive 3D Representation Learning Guided by Generative Pretraining
Viaarxiv icon

Autoencoders as Cross-Modal Teachers: Can Pretrained 2D Image Transformers Help 3D Representation Learning?

Dec 16, 2022
Runpei Dong, Zekun Qi, Linfeng Zhang, Junbo Zhang, Jianjian Sun, Zheng Ge, Li Yi, Kaisheng Ma

Figure 1 for Autoencoders as Cross-Modal Teachers: Can Pretrained 2D Image Transformers Help 3D Representation Learning?
Figure 2 for Autoencoders as Cross-Modal Teachers: Can Pretrained 2D Image Transformers Help 3D Representation Learning?
Figure 3 for Autoencoders as Cross-Modal Teachers: Can Pretrained 2D Image Transformers Help 3D Representation Learning?
Figure 4 for Autoencoders as Cross-Modal Teachers: Can Pretrained 2D Image Transformers Help 3D Representation Learning?
Viaarxiv icon