Research Experience

Projects

• Fact or Fairness? Identifying Over-Balanced Issues [Poster] Apr. 2024 - present
  Supervisor: Prof. Michael R. Lyu & Dr. Jen-tse Huang (ARISE@CUHK) Hong Kong SAR
  • Analyzed gender and racial biases in generative models using social demographic data, identifying over-balanced issues in content generation.
  • Investigated the trade-offs between fairness and accuracy in generative models, assessing how these factors influence model performance.
  • Developed quantitative metrics to evaluate fairness, improving bias detection and enhance fairness assessments in generative model outputs.
• Evaluation on the Vulnerability of Current Generative Models [Slides] Feb. 2024 - Jun. 2024
  Supervisor: Prof. Sabine Süsstrunk & Dr. Daichi Zhang (IVRL@EPFL) Lausanne, Switzerland
  • Assessed the defense mechanisms of current generative models against “jailbreak” attacks, revealing persistent security vulnerabilities.
  • Analyzed biases in diffusion models finding gender and racial biases in generated images, which calls for more diverse training datasets.
  • Evaluated fake image detectors uncovering demographic disparities in detection accuracy, highlighting the need for fairer training data.
• Efficient Video Analytics [Poster] Jun. 2023 - Sep. 2023
  Supervisor: Prof. Eric Lo (CPII@CUHK) Hong Kong SAR
  • Won the Best Project Award 2023 among 58 undergraduate projects.
  • Developed a multimodal AI system using CLIP and OWL-ViT models for text or image based lost-and-found queries at Hong Kong Airport.
  • Created custom datasets and evaluation standards to select the most accurate and efficient Zero-Shot Object Detection and NLP models.
  • Optimized frame detection with an algorithm that improved fault tolerance, allowing efficient, accurate search through large video datasets.
• Traffic Characteristics Analysis of the Network Apr. 2023 - Jun. 2023
  Supervisor: Prof. Tong Yang (DSE@PKU) Beijing, China
  • Developed an efficient platform to analyze campus network traffic, examining link, traffic, and packet levels to identify user preferences.
  • Mapped IP addresses to domain names using TCP/IP and DNS packet analysis, extracting data from pcap files to understand network patterns.
  • Analyzed campus network traffic to identify top domains and peak times with C++ and streaming algorithms for parsing and in-depth analysis.