Zekun Wu
Scholar

Zekun Wu

Google Scholar ID: xurQ_DgAAAAJ
Research Scientist, Holistic AI / PhD Student, University College London
Agentic AIResponsible AIBehavioural RobustnessExplainabilityInterpretability
Citations & Impact
All-time
Citations
119
 
H-index
6
 
i10-index
5
 
Publications
20
 
Co-authors
10
list available
Resume (English only)
Academic Achievements
  • - March 2025, Fortune featured Holistic AI's security audit of Anthropic's Claude 3.7
  • - February 2025, two research projects accepted at NAACL 2025: HyPA-RAG system and PEFT-driven personality manipulation
  • - January 2025, SAGED bias-benchmarking pipeline presented at COLING 2025
  • - December 2024, four research papers presented at NeurIPS 2024 workshops
Research Experience
  • - AI Research Scientist at Holistic AI
  • - Involved in policy research with OECD.AI
  • - Contributing to the EU AI Code of Practice
  • - Teaching at UCL
  • - Freelancing with SeeTalent to develop personality analysis tools
Education
  • - Degrees: Undergraduate, Master's
  • - University: University College London (UCL)
  • - Major: Computer Science
  • - Currently: Pursuing a part-time PhD at UCL's Department of Computer Science, focusing on responsible Agentic AI systems
Background
  • - Research Interests: Safety, Auditing, and Governance of Agentic AI
  • - Professional Field: Responsible AI Systems
  • - Introduction: Works as an AI Research Scientist at Holistic AI, ensuring AI systems behave properly. Also involved in policy research with OECD.AI, contributing to the EU AI Code of Practice, teaching at UCL, and freelancing with SeeTalent to develop personality analysis tools.
Miscellany
  • - Runs an online art studio, collecting and reselling jewelry, artwork, shoes, and designer goods