4 citations · 5 across the 18 of their papers we have counts for
20 papers
Compress to Remember: Learning Compact Memory via On-Policy Distillation for Long Video Generation
Xiaoyu Wu, Weihang Guo, Yifei Wang +3
Standard video generators do not natively compact historical context into reusable memory tokens. As generation continues, the growing history makes it increasingly difficult to re…
MarkTune: Improving the Quality-Detectability Trade-off in Model-Embedded LLM Watermarking
Yizhou Zhao, Zhiwei Steven Wu, Adam Block
Language model watermarking schemes fall into two broad categories: inference-time methods, which modify the decoding process at generation time, and model-embedded methods, which…
Measurement as Bricolage: Examining How Data Scientists Construct Target Variables for Predictive Modeling Tasks
Luke Guerdan, Devansh Saxena, Stevie Chancellor +2
Data scientists often formulate predictive modeling tasks involving fuzzy, hard-to-define concepts, such as the "authenticity" of student writing or the "healthcare need" of a pati…
Enhancing One-run Privacy Auditing with Quantile Regression-Based Membership Inference
Terrance Liu, Matteo Boglioni, Yiwei Fu +3
Differential privacy (DP) auditing aims to provide empirical lower bounds on the privacy guarantees of DP mechanisms like DP-SGD. While some existing techniques require many traini…
Membership Inference Attacks for Unseen Classes
Pratiksha Thaker, Neil Kale, Zhiwei Steven Wu +1
A key tool in developing safe AI models is \emph{data auditing}, i.e., using statistical tools to determine whether harmful content may have been used in the training data of a bla…
Generate-then-Verify: Reconstructing Data from Limited Published Statistics
Terrance Liu, Eileen Xiao, Adam Smith +2
We study the problem of reconstructing tabular data from aggregate statistics, in which the attacker aims to identify interesting claims about the sensitive data that can be verifi…
Winning the MIDST Challenge: New Membership Inference Attacks on Diffusion Models for Tabular Data Synthesis
Xiaoyu Wu, Yifei Pang, Terrance Liu +1
Tabular data synthesis using diffusion models has gained significant attention for its potential to balance data utility and privacy. However, existing privacy evaluations often re…
Improving the Convergence of Private Shuffled Gradient Methods with Public Data
Shuli Jiang, Pranay Sharma, Zhiwei Steven Wu +1
We consider the problem of differentially private (DP) convex empirical risk minimization (ERM). While the standard DP-SGD algorithm is theoretically well-established, practical im…
Inferentially-Private Private Information
Shuaiqi Wang, Shuran Zheng, Zinan Lin +2
Information disclosure can compromise privacy when revealed information is correlated with private information. We consider the notion of inferential privacy, which measures privac…
Position: LLM Unlearning Benchmarks are Weak Measures of Progress
Pratiksha Thaker, Shengyuan Hu, Neil Kale +3
Unlearning methods have the potential to improve the privacy and safety of large language models (LLMs) by removing sensitive or harmful information post hoc. The LLM unlearning re…
Multi-group Uncertainty Quantification for Long-form Text Generation
Terrance Liu, Zhiwei Steven Wu
While past works have shown how uncertainty quantification can be applied to large language model (LLM) outputs, the question of whether resulting uncertainty guarantees still hold…
Quantifying Privacy Risks of Public Statistics to Residents of Subsidized Housing
Ryan Steed, Diana Qing, Zhiwei Steven Wu
As the U.S. Census Bureau implements its controversial new disclosure avoidance system, researchers and policymakers debate the necessity of new privacy protections for public stat…
Reconciling Model Multiplicity for Downstream Decision Making
Ally Yalei Du, Dung Daniel Ngo, Zhiwei Steven Wu
We consider the problem of model multiplicity in downstream decision-making, a setting where two predictive models of equivalent accuracy cannot agree on the best-response action f…
Predictive Performance Comparison of Decision Policies Under Confounding
Luke Guerdan, Amanda Coston, Kenneth Holstein +1
Predictive models are often introduced to decision-making tasks under the rationale that they improve performance over an existing decision-making policy. However, it is challengin…
Guardrail Baselines for Unlearning in LLMs
Pratiksha Thaker, Yash Maurya, Shengyuan Hu +2
Recent work has demonstrated that finetuning is a promising approach to 'unlearn' concepts from large language models. However, finetuning can be expensive, as it requires both gen…
Differentially Private Bayesian Persuasion
Yuqi Pan, Zhiwei Steven Wu, Haifeng Xu +1
The tension between persuasion and privacy preservation is common in real-world settings. Online platforms should protect the privacy of web users whose data they collect, even as…
Oracle-Efficient Differentially Private Learning with Public Data
Adam Block, Mark Bun, Rathin Desai +2
Due to statistical lower bounds on the learnability of many function classes under privacy constraints, there has been recent interest in leveraging public data to improve the perf…
On the Benefits of Public Representations for Private Transfer Learning under Distribution Shift
Pratiksha Thaker, Amrith Setlur, Zhiwei Steven Wu +1
Public pretraining is a promising approach to improve differentially private model training. However, recent work has noted that many positive research results studying this paradi…
Ex-post Individually Rational Bayesian Persuasion
Jiahao Zhang, Shuran Zheng, Renato Paes Leme +1
In the Bayesian persuasion model, a sender can convince a receiver to choose an alternative action to the one originally preferred by the receiver. A crucial assumption in this mod…
Differentially Private SGD Without Clipping Bias: An Error-Feedback Approach
Xinwei Zhang, Zhiqi Bu, Zhiwei Steven Wu +1
Differentially Private Stochastic Gradient Descent with Gradient Clipping (DPSGD-GC) is a powerful tool for training deep learning models using sensitive data, providing both a sol…