CV

Curriculum vitae of Gang Xie, a master's student working on brain-inspired multimodal affective computing and reliable AI systems.

Contact Information

Name Gang Xie
Professional Title Master's Student in Computer Science
Email siegang@163.com

Professional Summary

Master’s student at Hangzhou Dianzi University researching brain-inspired multimodal affective computing and reliable AI systems.

Experience

  • 2025 - 2026

    Hangzhou, China

    Algorithm Engineer Intern
    Diling Zhiyuan Technology Co., Ltd.
    Contributed to algorithm development and engineering projects for intelligent content and information processing.

Education

  • 2024 - 2027

    Hangzhou, China

    Master's degree
    Hangzhou Dianzi University
    Computer Science and Technology
  • 2019 - 2023

    Kunming, China

    Bachelor's degree
    Kunming University of Science and Technology
    Intelligent Science and Technology

Projects

  • 2025 - 2026
    PROSE

    LLM reflection-driven multi-objective evolution for robust prompt optimization in multimodal content moderation.

    • Combined performance with a semantic abstraction proxy on a Pareto frontier to reduce case memorization and shortcut learning.
    • Built configuration-driven task adapters, heterogeneous model pools, concurrent evaluation, SQLite persistence, lineage tracking, and checkpoint recovery.
  • 2026 - 2026
    IntentSafe-CN

    Multi-layer diagnostic benchmark for intent-aware Chinese harmful-content detection.

    • Studied the contribution of surface signals, context, and communicative intent to downstream safety decisions.
    • Designed experiments, trained a lightweight intent model with LoRA, analyzed results, and prepared paper figures.

Publications

  • 2026
    From Enumeration to Principle: Robust Prompt Optimization via Semantic Reflection and Multi-Objective Evolution
    Findings of EMNLP 2026

    Accepted first-author paper introducing PROSE.

  • 2026
    NeuroSIFT: A Biologically-Inspired Framework with Explicit Signal-Noise Separation for Robust Multimodal Emotion Recognition
    ICASSP 2026

    Accepted first-author paper on robust multimodal emotion recognition.

Skills

Research (Advanced): Multimodal affective computing, multimodal emotion recognition, multimodal sentiment analysis, robustness, signal-noise separation
Large language models (Advanced): Prompt optimization, semantic reflection, multi-objective evolution, content safety, intent-aware detection
Engineering (Advanced): Python, PyTorch, Transformers, PEFT/LoRA, FastAPI, REST, SQLite, FAISS, Git

Awards

  • 2025
    Huawei Software Elite Challenge, Hangzhou-Xiamen Division, Second Prize
    Huawei
  • 2022
    Blue Bridge Cup, National Second Prize
    Blue Bridge Cup