default author photo

ShengZhi Shen

affiliation not provided to SSRN

  • affiliation not provided to SSRN

SCHOLARLY PAPERS

4

DOWNLOADS

92

TOTAL CITATIONS

0

Scholarly Papers (4)

Actions:
Email selected abstracts
1.

Enhancing Multimodal Large Language Models with Efficient Feature Alignment and Processing Using State Space Models

Number of pages: 20 Posted: 14 Jun 2025
Wenjun Huang, Jiakai Pan, Zhengzhuo Wang, Yuan Liu, Yuhe Wang, ShengZhi Shen and Jianguo Hu
affiliation not provided to SSRN, affiliation not provided to SSRN, affiliation not provided to SSRN, Chongqing University, affiliation not provided to SSRN, affiliation not provided to SSRN and affiliation not provided to SSRN
Downloads 39 (1,222,651)

Abstract:

Loading...

Mamba, Multimodal Models, Feature Alignment, Inference Speed

2.

Cluster-Masked Autoregressive Learning with xLSTM: A New Framework for General-Purpose Visual Representation

Number of pages: 28 Posted: 08 Nov 2025
Wenjun Huang, Yuan Liu, Yifei Xing, Pi Fang, Jiakai Pan, Yuhe Wang, ShengZhi Shen, Zhengzhuo Wang and Jianguo Hu
affiliation not provided to SSRN, affiliation not provided to SSRN, University of Chinese Academy of Sciences, South China Normal University, affiliation not provided to SSRN, affiliation not provided to SSRN, affiliation not provided to SSRN, affiliation not provided to SSRN and affiliation not provided to SSRN
Downloads 26 (1,385,945)

Abstract:

Loading...

Visual Representation Learning, Self-Supervised Learning, Autoregressive Pretraining, xLSTM, Computer Vision

3.

Shifting the Computational Burden: Efficient Long-Text MLLMs via LLM-Guided Visual Text Rendering

Number of pages: 21 Posted: 02 Feb 2026
Wenjun Huang, ShengZhi Shen, Yuan Liu, Pi Fang, Yong Wu, Yuhe Wang, Heng Zhang, Jiakai Pan, Zhengzhuo Wang and Jianguo Hu
affiliation not provided to SSRN, affiliation not provided to SSRN, affiliation not provided to SSRN, South China Normal University, South China Normal University, affiliation not provided to SSRN, South China Normal University, affiliation not provided to SSRN, affiliation not provided to SSRN and affiliation not provided to SSRN
Downloads 19 (1,481,165)

Abstract:

Loading...

Multimodal Learning, Visual Tokenization, LLM-Guided Training, Long-Context Modeling, Computational efficiency, Visual question answering

4.

Cluster-Masked Autoregressive Learning with xLSTM for Efficient Visual Representation

Number of pages: 34 Posted: 07 Apr 2026
Wenjun Huang, Yuan Liu, Yifei Xing, Jiakai Pan, Yuhe Wang, Pi Fang, ShengZhi Shen, Zhengzhuo Wang and Jianguo Hu
affiliation not provided to SSRN, affiliation not provided to SSRN, University of Chinese Academy of Sciences, affiliation not provided to SSRN, affiliation not provided to SSRN, South China Normal University, affiliation not provided to SSRN, affiliation not provided to SSRN and affiliation not provided to SSRN
Downloads 8 (1,561,096)

Abstract:

Loading...

Visual Representation Learning, Self-supervised learning, Autoregressive Pretraining, xLSTM, Cluster-Masked Modeling, medical image segmentation