Pratiksha Thaker

26 papers A* 4Misc 1Journal 16Unranked 5

Year	Rank	Type	Title / Venue / Authors
2025	J	`jnl`	BLUR: A Benchmark for LLM Unlearning Robust to Forget-Retain Overlap. CoRR Shengyuan Hu, Neil Kale, Pratiksha Thaker, Yiwei Fu, Steven Wu, Virginia Smith
2025	J	`jnl`	Enhancing One-run Privacy Auditing with Quantile Regression-Based Membership Inference. CoRR Terrance Liu, Matteo Boglioni, Yiwei Fu, Shengyuan Hu, Pratiksha Thaker, Zhiwei Steven Wu
2025	J	`jnl`	Generate-then-Verify: Reconstructing Data from Limited Published Statistics. CoRR Terrance Liu, Eileen Xiao, Pratiksha Thaker, Adam Smith, Zhiwei Steven Wu
2025	J	`jnl`	Lower Bounds for Public-Private Learning under Distribution Shift. CoRR Amrith Setlur, Pratiksha Thaker, Jonathan Ullman
2025	J	`jnl`	Membership Inference Attacks for Unseen Classes. CoRR Pratiksha Thaker, Neil Kale, Zhiwei Steven Wu, Virginia Smith
2025	J	`jnl`	PARALLELPROMPT: Extracting Parallelism from Large Language Model Queries. CoRR Steven Kolawole, Keshav Santhanam, Virginia Smith, Pratiksha Thaker
2025	—	`conf`	Position: LLM Unlearning Benchmarks are Weak Measures of Progress. SaTML Pratiksha Thaker, Shengyuan Hu, Neil Kale, Yash Maurya, Zhiwei Steven Wu, Virginia Smith
2024	—	`conf`	ALTO: An Efficient Network Orchestrator for Compound AI Systems. EuroMLSys@EuroSys Keshav Santhanam, Deepti Raghavan, Muhammad Shahir Rahman, Thejas Venkatesh, Neha Kunjal, Pratiksha Thaker, Philip Alexander Levis, Matei Zaharia
2024	J	`jnl`	ALTO: An Efficient Network Orchestrator for Compound AI Systems. CoRR Keshav Santhanam, Deepti Raghavan, Muhammad Shahir Rahman, Thejas Venkatesh, Neha Kunjal, Pratiksha Thaker, Philip Alexander Levis, Matei Zaharia
2024	J	`jnl`	Guardrail Baselines for Unlearning in LLMs. CoRR Pratiksha Thaker, Yash Maurya, Virginia Smith
2024	A*	`conf`	On the Benefits of Public Representations for Private Transfer Learning under Distribution Shift. NeurIPS Pratiksha Thaker, Amrith Setlur, Steven Z. Wu, Virginia Smith
2024	J	`jnl`	Position: LLM Unlearning Benchmarks are Weak Measures of Progress. CoRR Pratiksha Thaker, Shengyuan Hu, Neil Kale, Yash Maurya, Zhiwei Steven Wu, Virginia Smith
2023	A*	`conf`	Congestion Control Safety via Comparative Statics. INFOCOM Pratiksha Thaker, Matei Zaharia, Tatsunori Hashimoto
2023	A*	`conf`	Cornflakes: Zero-Copy Serialization for Microsecond-Scale Networking. SOSP Deepti Raghavan, Shreya Ravi, Gina Yuan, Pratiksha Thaker, Sanjari Srivastava, Micah Murray, Pedro Henrique Penna, Amy Ousterhout, Philip Alexander Levis, Matei Zaharia, Irene Zhang
2023	J	`jnl`	Leveraging Public Representations for Private Transfer Learning. CoRR Pratiksha Thaker, Amrith Setlur, Zhiwei Steven Wu, Virginia Smith
2023	—	`conf`	On Noisy Evaluation in Federated Hyperparameter Tuning. MLSys Kevin Kuo, Pratiksha Thaker, Mikhail Khodak, John Nguyen, Daniel Jiang, Ameet Talwalkar, Virginia Smith
2022	J	`jnl`	On Noisy Evaluation in Federated Hyperparameter Tuning. CoRR Kevin Kuo, Pratiksha Thaker, Mikhail Khodak, John Nguyen, Daniel Jiang, Ameet Talwalkar, Virginia Smith
2022	J	`jnl`	Overlook: Differentially Private Exploratory Visualization for Big Data. J. Priv. Confidentiality Mihai Budiu, Pratiksha Thaker, Parikshit Gopalan, Udi Wieder, Matei Zaharia
2021	—	`conf`	Clamor: Extending Functional Cluster Computing Frameworks with Fine-Grained Remote Memory Access. SoCC Pratiksha Thaker, Hudson Ayers, Deepti Raghavan, Ning Niu, Philip Alexander Levis, Matei Zaharia
2021	Misc	`conf`	Don't Hate the Player, Hate the Game: Safety and Utility in Multi-Agent Congestion Control. HotNets Pratiksha Thaker, Matei Zaharia, Tatsunori Hashimoto
2020	J	`jnl`	Overlook: Differentially Private Exploratory Visualization for Big Data. CoRR Pratiksha Thaker, Mihai Budiu, Parikshit Gopalan, Udi Wieder, Matei Zaharia
2018	J	`jnl`	Evaluating End-to-End Optimization for Data Analytics Applications in Weld. Proc. VLDB Endow. Shoumik Palkar, James Thomas, Deepak Narayanan, Pratiksha Thaker, Rahul Palamuttam, Parimarjan Negi, Anil Shanbhag, Malte Schwarzkopf, Holger Pirk, Saman P. Amarasinghe, Samuel Madden, Matei Zaharia
2014	A*	`conf`	An experimental study of the learnability of congestion control. SIGCOMM Anirudh Sivaraman, Keith Winstein, Pratiksha Thaker, Hari Balakrishnan
2014	J	`jnl`	Learning perceptually grounded word meanings from unaligned parallel data. Mach. Learn. Stefanie Tellex, Pratiksha Thaker, Joshua Mason Joseph, Nicholas Roy
2013	J	`jnl`	Clarifying commands with information-theoretic human-robot dialog. J. Hum. Robot Interact. Robin Deits, Stefanie Tellex, Pratiksha Thaker, Dimitar Simeonov, Thomas Kollar, Nicholas Roy
2012	—	`conf`	Toward Information Theoretic Human-Robot Dialog. Robotics: Science and Systems Stefanie Tellex, Pratiksha Thaker, Robin Deits, Thomas Kollar, Nicholas Roy