Pankayaraj Pathmanathan
Scholar

Pankayaraj Pathmanathan

Google Scholar ID: G_gAOpwAAAAJ
University of Maryland College Park
LLM securityRetrieval Augmented GenerationPoisoningLLM AlignmentReinforcement Learning
Citations & Impact
All-time
Citations
80
 
H-index
4
 
i10-index
3
 
Publications
12
 
Co-authors
0
 
Resume (English only)
Academic Achievements
  • Publications: 'Robust Reward modeling for LLM alignment' (EMNLP 2025); 'RAGPart & RAGMask: Retrieval-Stage Defenses Against Corpus Poisoning in Retrieval-Augmented Generation' (COLM 2025). Won the best final year project research thesis award.
Research Experience
  • Worked as a research engineer at CARE AI Lab, Singapore Management University, under Prof. Pradeep Varakantham; Research Assistant and intern at Sri Lanka Technological Campus, supervised by Prof. D.H.S. Maithripala; Will be joining Netflix Research USA as a machine learning research intern in summer 2025.
Education
  • PhD: Department of Computer Science, University of Maryland, Advisor: Prof. Furong Huang; BSc: Computer Engineering, University of Peradeniya, Advisor: Prof. D.H.S. Maithripala
Background
  • Research interests: Reinforcement Learning (RL), constrained RL, continual RL, multi-agent RL, Bayesian RL, bandits. Currently interested in LLM and RLHF poisoning.
Miscellany
  • Personal website provides contact information and links to CVs.
Co-authors
0 total
Co-authors: 0 (list not available)