Meng Wang

RA at CAIR-CAS. Interested in LLM safety & alignment.

prof_pic.avif

Hi! I am Meng, currently a research assistant working with Prof. Fei Zhu at CAIR-CAS. I obtained my Master’s in Computer Science from University of Rochester in 2025 and BSc in Computer Science from Hong Kong PolyU in 2023. My research interests include LLM agents and continual learning.

news

Dec 17, 2025 [ArXiv] One first author paper, the first vision-text compression benchmark: VTCBench.
Oct 01, 2025 [ArXiv] Our work on Continual Learning for Multimodal Large Language Models.
Jul 21, 2025 Joined CAIR-CAS as an RA!
Apr 19, 2025 [CISA’25] One first-author paper on computational imaging accepted to CISA 2025 Oral!
Feb 22, 2025 [CHI’25] One first-author paper on LLM agent got rejected! Redoing evaluation…

selected publications

  1. VTCBench: Can Vision-Language Models Understand Long Context with Vision-Text Compression?
    Hongbo Zhao, Meng Wang, Fei Zhu, and 5 more authors
    Dec 2025
    DOI
  2. MRLoRA_preview.jpg
    MLLM-CL: Continual Learning for Multimodal Large Language Models
    Hongbo Zhao, Fei Zhu, Haiyang Guo, and 4 more authors
    Oct 2025
    DOI
  3. iRADNet_preview.png
    Deep Algorithm Unrolling for Seismic Migration
    Meng Wang and Tolulope Olugboji
    In 2025 IEEE Conference on Computational Imaging Using Synthetic Apertures (CISA), Jun 2025
    DOI