Pratiksha Thaker

Carnegie Mellon University — United States

Computer Science · Physical Sciences

9h-index477citations30works0.82yr avg
OpenAlex↗ORCID↗Google Scholar↗Lab page↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Privacy-Preserving Technologies in Data(5), Topic Modeling(4), Network Traffic and Congestion Control(4), Natural Language Processing Techniques(4), Cloud Computing and Resource Management(3)

Publications30 total

Lower Bounds for Public-Private Learning under Distribution Shift
arXiv (Cornell University)·2025Open Access
Enhancing One-run Privacy Auditing with Quantile Regression-Based Membership Inference
arXiv (Cornell University)·2025Open Access
Membership Inference Attacks for Unseen Classes
arXiv (Cornell University)·2025Open Access
Generate-then-Verify: Reconstructing Data from Limited Published Statistics
arXiv (Cornell University)·2025Open Access
Position: LLM Unlearning Benchmarks are Weak Measures of Progress
2025· 1 cited
Position: LLM Unlearning Benchmarks are Weak Measures of Progress
arXiv (Cornell University)·2024· 1 citedOpen Access
ALTO
2024· 3 citedOpen Access
Alto: Orchestrating Distributed Compound AI Systems with Nested Ancestry
arXiv (Cornell University)·2024Open Access
Guardrail Baselines for Unlearning in LLMs
arXiv (Cornell University)·2024· 3 citedOpen Access
On the Benefits of Public Representations for Private Transfer Learning under Distribution Shift
2024
On the Benefits of Public Representations for Private Transfer Learning under Distribution Shift
arXiv (Cornell University)·2023Open Access
Cornflakes: Zero-Copy Serialization for Microsecond-Scale Networking
2023· 17 cited
Congestion Control Safety via Comparative Statics
2023· 1 cited
On Noisy Evaluation in Federated Hyperparameter Tuning
arXiv (Cornell University)·2022Open Access
Overlook: Differentially Private Exploratory Visualization for Big Data
Journal of Privacy and Confidentiality·2022Open Access
Don't Hate the Player, Hate the Game
2021· 4 cited
Clamor
2021· 1 citedOpen Access
Overlook: Differentially Private Exploratory Visualization for Big Data
arXiv (Cornell University)·2020· 8 citedOpen Access
Evaluating end-to-end optimization for data analytics applications in weld
Proceedings of the VLDB Endowment·2018· 69 citedOpen Access
Semantics-Aware Program Sampling
2017
page 1 of 2Next →

Frequent Co-authors

Matei Zaharia(9), Virginia Smith(7), Zhiwei Steven Wu(7), Stefanie Tellex(6), Nicholas Roy(6), Robin Deits(4), Thomas Kollar(4), Deepti Raghavan(4), Philip Levis(4), Dimitar Simeonov(3), Yash Maurya(3), Shengyuan Hu(3), Neil Kale(3), Amrith Setlur(3), Anirudh Sivaraman(2), Keith Winstein(2), Hari Balakrishnan(2), Mihai Budiu(2), Parikshit Gopalan(2), Udi Wieder(2)