Research

My research lies at the intersection of data science, animal welfare science, and AI bias. My previous research focused on advancing automated methods for behavior monitoring, redefining dominance hierarchy calculations, creating novel approaches for lameness detection, and understanding behavioral baselines for individual animals from sensor data. Currently, my work explores the representation bias of livestock farming in generative AI.

Publications

The erasure of intensive livestock farming in text-to-image generative AI
Sheng, K., Tuyttens, F.A.M., von Keyserlingk, M.A.G.
ACM Conference on Fairness, Accountability, and Transparency (FAccT) 2025
Paper Code
Redefining Lameness Assessment: Constructing Lameness Hierarchy using Crowd-Sourced Data
Sheng, K., Foris, B., von Keyserlingk, M.A.G., Timbers, T.A., Cabrera, V., Weary, D.M.
Computers and Electronics in Agriculture (2025)
Paper Code
Redefining dominance calculation: Increased competition flattens the dominance hierarchy in dairy cows
Sheng, K., Foris, B., Krahn, J., Weary, D.M., von Keyserlingk, M.A.G.
Journal of Dairy Science, 107(9): 7286-7298 (2024)
Paper Code
Crowd sourcing remote comparative lameness assessments for dairy cattle
Sheng, K., Foris, B., von Keyserlingk, M.A.G., Gardenier, J., Clark, C., Weary, D.M.
Journal of Dairy Science, 106(8): 5715-5722 (2023)
Paper
Automated, longitudinal measures of drinking behavior provide insights into the social hierarchy in dairy cows
Foris, B., Vandresen, B., Sheng, K., Krahn, J., Weary, D.M., von Keyserlingk, M.A.G.
JDS Communications, 5(5): 411-415 (2024)
Paper
Effects of group size on agonistic interactions in dairy cows: a descriptive study
Krahn, J., Foris, B., Sheng, K., Weary, D.M., von Keyserlingk, M.A.G.
Animal, 18(3): 101083 (2024)
Paper

Research Interests

Current Focus

  • AI bias and representation in animal welfare contexts
  • Automated farm animal behaviour andwelfaremonitoring
  • Reproducible research methodologies

Technical Expertise

  • The application of AI agents and generative AI models (e.g., large language models and text-to-image models)
  • Machine learning and computer vision for animal behavior analysis
  • Crowdsourcing and human-AI collaboration
  • Reproducible data science workflow and software development (R, Python)