ALU uses public data to suppress unlearning cost quadratically while characterizing distribution mismatch effects, enabling mass unlearning with maintained utility.
https://arxiv.org/pdf/2305.08846
6 Pith papers cite this work, alongside 9 external citations. Polarity classification is still indexing.
6
Pith papers citing it
9
external citations · Pith
abstract
We propose a scheme for auditing differentially private machine learning systems with a single training run. This exploits the parallelism of being able to add or remove multiple training examples independently. We analyze this using the connection between differential privacy and statistical generalization, which avoids the cost of group privacy. Our auditing scheme requires minimal assumptions about the algorithm and can be applied in the black-box or white-box setting.
representative citing papers
Min-K% Prob detects pretraining data in LLMs by flagging outlier low-probability words in text, achieving 7.4% better performance than prior methods on the new WIKIMIA benchmark.
An auditor based on membership inference attacks computes valid lower bounds on the unlearning parameter ε, empirically separating certified unlearning methods (small bounds) from heuristic ones (large bounds).
DP-SGD with expected or batch averaging (EASGM or ASGM) has weaker privacy guarantees than the standard subsampled Gaussian mechanism analysis, confirmed by theoretical re-analysis and audits of libraries including Opacus.
TOFU is a new benchmark with synthetic profiles and metrics demonstrating that existing unlearning algorithms for LLMs fail to achieve effective forgetting of targeted information.
A data-centric survey finds that only information-flow control covers compositional and cross-session leakage in LLM agents and that no single benchmark tests an agent across all its data surfaces under one policy.
citing papers explorer
-
Unlearning with Asymmetric Sources: Improved Unlearning-Utility Trade-off with Public Data
ALU uses public data to suppress unlearning cost quadratically while characterizing distribution mismatch effects, enabling mass unlearning with maintained utility.
-
Detecting Pretraining Data from Large Language Models
Min-K% Prob detects pretraining data in LLMs by flagging outlier low-probability words in text, achieving 7.4% better performance than prior methods on the new WIKIMIA benchmark.
-
Auditing of Unlearning Algorithms
An auditor based on membership inference attacks computes valid lower bounds on the unlearning parameter ε, empirically separating certified unlearning methods (small bounds) from heuristic ones (large bounds).
-
Rethinking the Security of DP-SGD: A Corrected Analysis of Differentially Private Machine Learning
DP-SGD with expected or batch averaging (EASGM or ASGM) has weaker privacy guarantees than the standard subsampled Gaussian mechanism analysis, confirmed by theoretical re-analysis and audits of libraries including Opacus.
-
TOFU: A Task of Fictitious Unlearning for LLMs
TOFU is a new benchmark with synthetic profiles and metrics demonstrating that existing unlearning algorithms for LLMs fail to achieve effective forgetting of targeted information.
-
Agents That Know Too Much: A Data-Centric Survey of Privacy in LLM Agents
A data-centric survey finds that only information-flow control covers compositional and cross-session leakage in LLM agents and that no single benchmark tests an agent across all its data surfaces under one policy.