Research Interests

• LLM Agents
• Physics-Inspired Oscillatory Neural Networks
• Diffusion Language Models
• Graph Neural Networks
• AI Security

Publications

• OpenFinGym: A Verifiable Multi-Task Gym Environment for Evaluating Quant Agents
  K Zhang*, W Ge*, L Jiang, W Yang, J Langham-Lopez, J Yu, L Szpruch, H Ni
  arXiv preprint arXiv:2606.26350, 2026
  [arXiv]
• MOSAIC: Modular Orchestration for Structured Agentic Intelligence and Composition
  Y Bao, X Xi, X Liu, W Ge, L Jiang, K Zhang, R Khraishi, Y Ang, AKH Tung, L Szpruch, H Ni
  arXiv preprint arXiv:2606.00708, 2026
  [arXiv]
• Stuart-Landau Oscillatory Graph Neural Network
  K Zhang, DN Reynolds, P Deidda, F Tudisco
  WWW 2026
  [arXiv] [Code]
• Rethinking the Vulnerability of Concept Erasure and a New Method
  AD Richardson*, K Zhang*, L Beerens, D Chen
  arXiv preprint arXiv:2502.17537, 2025
  [arXiv]
• Are We Measuring Oversmoothing in Graph Neural Networks Correctly?
  K Zhang*, P Deidda*, D Higham, F Tudisco
  ICLR 2026
  [arXiv]
• Adversarial Attacks on Convolutional Neural Networks with General Purpose Optimization Software
  K Zhang, D Zorba, A Meda, L Beerens, D Higham
  MadeAI, 2025
• Optical Neural Networks Based on Perovskite Solar Cells
  K Zhang, J Harwell, D Pierangeli, C Conti, A Di Falco
  Photonics Research 13 (2), 382-386, 2025
  [Paper]
• A Low-Rank Perspective on Oversmoothing in Graph Neural Networks
  K Zhang, P Deidda, D Higham, F Tudisco
  AAAI 2025 Workshop CoLoRAI, 2025
  [PDF]