Author

Pankayaraj Pathmanathan

Observed name: Pathmanathan, Pankayaraj

Institution: University of Maryland, College Park

Observed institution: University of Maryland, College Park

OpenAlex ID: A5113258470

Can Watermarking Large Language Models Prevent Copyrighted Text Generation and Hide Training Data?

2025 · Michael-Andrei Panaitescu-Liess, Zora Che, et al. · Proceedings of the AAAI Conference on Artificial Intelligence · 3

PDF

Is poisoning a real threat to LLM alignment? Maybe more so than you think

2024 · Pathmanathan, Pankayaraj, Chakraborty, Souradip, et al. · arXiv (Cornell University) · 3

PDF

Is Poisoning a Real Threat to DPO? Maybe More So Than You Think

2025 · Pankayaraj Pathmanathan, Souradip Chakraborty, et al. · Proceedings of the AAAI Conference on Artificial Intelligence · 2

PDF

AdvBDGen: A Robust Framework for Generating Adaptive and Stealthy Backdoors in LLM Alignment

2026 · Pankayaraj Pathmanathan, Udari Madhushani Sehwag, et al. · Proceedings of the AAAI Conference on Artificial Intelligence · 1

PDF

Compliance2LoRA: Personalizable On-Demand Safety Alignment on Arbitrary Policy Subsets via Hypernetwork-Generated LoRA Adapters

2026 · Pathmanathan, Pankayaraj, Huang, Furong · arXiv (Cornell University)

PDF

Teach a Reward Model to Correct Itself: Reward Guided Adversarial Failure Discovery for Robust Reward Modeling

2026 · Association for Computational Linguistics 2026, Huang, Furong, et al. · Underline Science Inc.

Deliberative Alignment is Deep, but Uncertainty Remains: Inference time safety improvement in reasoning via attribution of unsafe behavior to base model

2026 · Pathmanathan, Pankayaraj, Huang, Furong · arXiv (Cornell University)

PDF

AdvBDGen: A Robust Framework for Generating Adaptive and Stealthy Backdoors in LLM Alignment

2026 · Association for Artificial Intelligence 2026, Chiang, Cho-Yu Jason, et al. · Underline Science Inc.

Teach a Reward Model to Correct Itself: Reward Guided Adversarial Failure Discovery for Robust Reward Modeling

2026 · Pankayaraj Pathmanathan, Furong Huang · Proceedings of the conference - Association for Computational Linguistics. Meeting

PDF

RAGPart & RAGMask: Retrieval-Stage Defenses Against Corpus Poisoning in Retrieval-Augmented Generation

2025 · Pathmanathan, Pankayaraj, Panaitescu-Liess, Michael-Andrei, et al. · arXiv (Cornell University)

PDF

PoisonedParrot: Subtle Data Poisoning Attacks to Elicit Copyright-Infringing Content from Large Language Models

2025 · Panaitescu-Liess, Michael-Andrei, Pathmanathan, Pankayaraj, et al. · arXiv (Cornell University)

PDF

PoisonedParrot: Subtle Data Poisoning Attacks to Elicit Copyright-Infringing Content from Large Language Models

2025 · Michael-Andrei Panaitescu-Liess, Pankayaraj Pathmanathan, et al. · Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT)

PDF

AdvBDGen: Adversarially Fortified Prompt-Specific Fuzzy Backdoor Generator Against LLM Alignment

2024 · Pathmanathan, Pankayaraj, Sehwag, Udari Madhushani, et al. · arXiv (Cornell University)

PDF

Works count: 15

Citation count: 11

H-index: 2

I10-index: