Zhengqin Lai

Ph.D. student, Computer Science — Harbin Institute of Technology (Shenzhen)

quenlenu@gmail.com Google Scholar
Portrait of Zhengqin Lai

About

I am a second-year CS Ph.D. student at Harbin Institute of Technology (Shenzhen). My research interests focus on lifelong learning for embodied intelligence.

Timeline

  • 2024 — Present

    Ph.D. Student

    Computer Science

    Harbin Institute of Technology (Shenzhen)

  • 2020 — 2024

    Undergraduate Student

    Computer Science and Technology

    Harbin Institute of Technology

Research

I pursue a single thread of lifelong and incremental learning across multimodal perception and affective computing. My recent work spans open-vocabulary multimodal emotion recognition (Agent-MER), incremental protocols for micro-expression recognition, robust multimodal modeling for sentiment understanding, and data-efficient “free-lunch” enhancements for crowd counting. My long-term goal is to build embodied agents that acquire skills continuously in real environments.

Selected Publications

  • Incremental Micro-Expression Recognition: A Benchmark

    Zhengqin Lai, Xiaopeng Hong*, Yabin Wang, Xiaobai Li

    IEEE Transactions on Affective Computing (TAFFC), 2026

    TAFFC · DOI

  • Agent-MER: A Cognitive Agent with Hierarchical Deliberation for Open-Vocabulary Multimodal Emotion Recognition

    Zhengqin Lai, Zhilin Zhu, Xiaopeng Hong*, Yaowei Wang

    Proceedings of the ACM International Conference on Multimedia (ACM MM), 2025

    ACM DL · DOI

  • Multimodal Blockwise Transformer for Robust Sentiment Recognition

    Zhengqin Lai, Xiaopeng Hong*, Yabin Wang

    Proceedings of the 2nd International Workshop on Multimodal and Responsible Affective Computing (MRAC), 2024

    ACM DL · DOI

  • SER: Learning to Ground Video Reasoning with Semantic Evidence Rewards

    Sheng Xia†, Zhengqin Lai†, Tianxiang Jiang, Kanghui Tian, Shoujun Zhou, Bin Li, Yi Wang († equal contribution)

    arXiv preprint arXiv:2606.24726, 2026

    arXiv · Google Scholar

  • Free Lunch Enhancements for Multi-modal Crowd Counting

    Haoliang Meng, Xiaopeng Hong*, Zhengqin Lai, Miao Shang

    Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2025

    CVF · DOI · Code

Honors & Awards

  • IJCAI @ MER 2024 — NOISE Track, Rank 3
  • ACM MM @ MER 2025 — FG Track, Rank 1
  • ACM MM @ EgoLink 2026 — Track 1, Rank 2
  • ACM MM @ PCBA 2026 — Rank 3
  • CSIG CEI 2025 — Best Poster Award

Links