팔로우
Aounon Kumar
Aounon Kumar
Research Associate, Harvard University
hbs.edu의 이메일 확인됨 - 홈페이지
제목
인용
인용
연도
Can AI-generated text be reliably detected?
VS Sadasivan, A Kumar, S Balasubramanian, W Wang, S Feizi
arXiv preprint arXiv:2303.11156, 2023
4702023
Certifying llm safety against adversarial prompting
A Kumar, C Agarwal, S Srinivas, AJ Li, S Feizi, H Lakkaraju
arXiv preprint arXiv:2309.02705, 2023
1842023
On the cost of essentially fair clusterings
IO Bercea, M Groß, S Khuller, A Kumar, C Rösner, DR Schmidt, ...
arXiv preprint arXiv:1811.10319, 2018
1332018
Curse of dimensionality on randomized smoothing for certifiable robustness
A Kumar, A Levine, T Goldstein, S Feizi
International Conference on Machine Learning, 5458-5467, 2020
1052020
Detection as regression: Certified object detection with median smoothing
P Chiang, M Curry, A Abdelkader, A Kumar, J Dickerson, T Goldstein
Advances in Neural Information Processing Systems 33, 1275-1286, 2020
762020
Policy smoothing for provably robust reinforcement learning
A Kumar, A Levine, S Feizi
arXiv preprint arXiv:2106.11420, 2021
622021
Robustness of ai-image detectors: Fundamental limits and practical attacks
M Saberi, VS Sadasivan, K Rezaei, A Kumar, A Chegini, W Wang, S Feizi
arXiv preprint arXiv:2310.00076, 2023
502023
Certifying confidence via randomized smoothing
A Kumar, A Levine, S Feizi, T Goldstein
Advances in Neural Information Processing Systems 33, 5165-5177, 2020
452020
Towards safe and aligned large language models for medicine
T Han, A Kumar, C Agarwal, H Lakkaraju
arXiv e-prints, arXiv: 2403.03744, 2024
26*2024
Center smoothing: Certified robustness for networks with structured outputs
A Kumar, T Goldstein
Advances in Neural Information Processing Systems 34, 5560-5575, 2021
26*2021
Manipulating large language models to increase product visibility
A Kumar, H Lakkaraju
arXiv preprint arXiv:2404.07981, 2024
15*2024
Tight second-order certificates for randomized smoothing
A Levine, A Kumar, T Goldstein, S Feizi
arXiv preprint arXiv:2010.10549, 2020
152020
Certifying model accuracy under distribution shifts
A Kumar, A Levine, T Goldstein, S Feizi
arXiv preprint arXiv:2201.12440, 2022
132022
Provable robustness against wasserstein distribution shifts via input randomization
A Kumar, A Levine, T Goldstein, S Feizi
ICLR 2023, 2023
62023
Capacitated k-center problem with vertex weights
A Kumar
36th IARCS Annual Conference on Foundations of Software Technology and …, 2016
52016
Generalizing Trust: Weak-to-Strong Trustworthiness in Language Models
M Pawelczyk, L Sun, Z Qi, A Kumar, H Lakkaraju
arXiv preprint arXiv:2501.00418, 2024
12024
Detecting LLM-Written Peer Reviews
V Rao, A Kumar, H Lakkaraju, NB Shah
arXiv preprint arXiv:2503.15772, 2025
2025
Provable Robustness for Streaming Models with a Sliding Window
A Kumar, VS Sadasivan, S Feizi
arXiv preprint arXiv:2303.16308, 2023
2023
Extending the Scope of Provable Adversarial Robustness in Machine Learning
A Kumar
University of Maryland, College Park, 2023
2023
Weak-to-Strong Trustworthiness: Eliciting Trustworthiness with Weak Supervision
M Pawelczyk, L Sun, Z Qi, A Kumar, H Lakkaraju
현재 시스템이 작동되지 않습니다. 나중에 다시 시도해 주세요.
학술자료 1–20