Get started
Author

Ashwinee Panda

FetchSGD: Communication-Efficient Federated Learning with Sketching
2020 · Rothchild, Daniel, Panda, Ashwinee, et al. · arXiv (Cornell University) · 234
PDF
Visual Adversarial Examples Jailbreak Aligned Large Language Models
2024 · Xiangyu Qi, Kaixuan Huang, et al. · Proceedings of the AAAI Conference on Artificial Intelligence · 150
PDF
Neurotoxin: Durable Backdoors in Federated Learning
2022 · Zhang, Zhengming, Panda, Ashwinee, et al. · arXiv (Cornell University) · 88
PDF
SparseFed: Mitigating Model Poisoning Attacks in Federated Learning with Sparsification
2021 · Panda, Ashwinee, Mahloujifar, Saeed, et al. · arXiv (Cornell University) · 54
PDF
Differentially Private Image Classification by Learning Priors from Random Processes
2023 · Tang, Xinyu, Panda, Ashwinee, et al. · arXiv (Cornell University) · 6
PDF
Privacy-Preserving In-Context Learning for Large Language Models
2023 · Wu, Tong, Panda, Ashwinee, et al. · arXiv (Cornell University) · 6
PDF
Safety Alignment Should Be Made More Than Just a Few Tokens Deep
2024 · Qi, Xiangyu, Panda, Ashwinee, et al. · arXiv (Cornell University) · 5
PDF
Teach LLMs to Phish: Stealing Private Information from Language Models
2024 · Panda, Ashwinee, Choquette-Choo, Christopher A., et al. · arXiv (Cornell University) · 5
PDF
DynaGuard: A Dynamic Guardian Model With User-Defined Policies
2025 · Hoover, Monte, Baherwani, Vatsal, et al. · arXiv (Cornell University) · 1
PDF
Differentially Private Image Classification by Learning Priors from Random Processes
2025 · Xinyu Tang, Ashwinee Panda, et al. · Journal of Privacy and Confidentiality · 1
Dense Backpropagation Improves Training for Sparse Mixture-of-Experts
2025 · Ashwinee Panda, Vatsal Baherwani, et al. · neural information processing systems · 1
PDF
Refusal Tokens: A Simple Way to Calibrate Refusals in Large Language Models
2024 · Jain, Neel, Shrivastava, Aditya, et al. · arXiv (Cornell University) · 1
PDF
A New Linear Scaling Rule for Private Adaptive Hyperparameter Optimization
2022 · Panda, Ashwinee, Tang, Xinyu, et al. · arXiv (Cornell University) · 1
PDF
Not All LLM Reasoning is Visible in the Chain-of-Thought
2026 · Baherwani, Vatsal, Goldstein, Tom, et al. · arXiv (Cornell University)
PDF
Emergent Misalignment via In-Context Learning: Narrow in-context examples can produce broadly misaligned LLMs
2026 · Association for Computational Linguistics 2026, Afonin, Nikita, et al. · Underline Science Inc.
From Refusal Tokens to Refusal Control: Discovering and Steering Category-Specific Refusal Directions
2026 · Alagharu, Rishab, Singh, Ishneet Sukhvinder, et al. · arXiv (Cornell University)
PDF
Speculating Experts Accelerates Inference for Mixture-of-Experts
2026 · Madan, Vivan, Singhania, Prajwal, et al. · arXiv (Cornell University)
PDF
@GrokSet: multi-party Human-LLM Interactions in Social Media
2026 · Migliarini, Matteo, Ercevik, Berat, et al. · arXiv (Cornell University)
PDF
Multi-Token Prediction via Self-Distillation
2026 · Kirchenbauer, John, Hans, Abhimanyu, et al. · arXiv (Cornell University)
PDF
A Few Bad Neurons: Isolating and Surgically Correcting Sycophancy
2026 · O'Brien, Claire, Seto, Jessica, et al. · arXiv (Cornell University)
PDF
Emergent Misalignment via In-Context Learning: Narrow in-context examples can produce broadly misaligned LLMs
2026 · Nikita Afonin, Nikita Andriianov, et al. · Proceedings of the conference - Association for Computational Linguistics. Meeting
PDF
Peek-a-Boo Reasoning: Contrastive Region Masking in MLLMs
2025 · Chaturvedi, Isha, Nair, Anjana, et al. · arXiv (Cornell University)
PDF
Modeling and Predicting Multi-Turn Answer Instability in Large Language Models
2025 · He, Jiahang, Ramachandran, Rishi, et al. · arXiv (Cornell University)
PDF
SALT: Steering Activations towards Leakage-free Thinking in Chain of Thought
2025 · Batra, Shourya, Tillman, Pierce, et al. · arXiv (Cornell University)
PDF
Alignment-Constrained Dynamic Pruning for LLMs: Identifying and Preserving Alignment-Critical Circuits
2025 · Patel, Dev, Gervacio, Gabrielle, et al. · arXiv (Cornell University)
PDF
Works count: 39
Citation count: 554
H-index: 6
I10-index: 4
Year
Topic
Type