REVIEW 7 cited by
Quantifying Privacy Risks of Masked Language Models Using Membership Inference Attacks
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
The wide adoption and application of Masked language models~(MLMs) on sensitive data (from legal to medical) necessitates a thorough quantitative investigation into their privacy vulnerabilities -- to what extent do MLMs leak information about their training data? Prior attempts at measuring leakage of MLMs via membership inference attacks have been inconclusive, implying the potential robustness of MLMs to privacy attacks. In this work, we posit that prior attempts were inconclusive because they based their attack solely on the MLM's model score. We devise a stronger membership inference attack based on likelihood ratio hypothesis testing that involves an additional reference MLM to more accurately quantify the privacy risks of memorization in MLMs. We show that masked language models are extremely susceptible to likelihood ratio membership inference attacks: Our empirical results, on models trained on medical notes, show that our attack improves the AUC of prior membership inference attacks from 0.66 to an alarmingly high 0.90 level, with a significant improvement in the low-error region: at 1% false positive rate, our attack is 51X more powerful than prior work.
Forward citations
Cited by 7 Pith papers
-
In-Context Probing for Membership Inference in Fine-Tuned Language Models
ICP-MIA infers membership in fine-tuned LLMs by measuring confidence improvement under in-context probes, beating prior black-box attacks at low false-positive rates.
-
LLM in the Middle: A Systematic Review of Threats and Mitigations to Real-World LLM-based Systems
A systematic review that categorizes LLM threats, severity scores, and mitigations across development and operation life cycles and multiple deployment scenarios.
-
Ensembling Membership Inference Attacks Against Tabular Generative Models
No single membership inference attack dominates across tabular generative models, and unsupervised ensembles of attacks achieve better average rankings.
-
Evaluating the Dynamics of Membership Privacy in Deep Learning
Per-sample membership vulnerability is established early in training, especially for hard-to-learn examples, and can be tracked on an FPR-TPR plane.
-
Identifying Pre-training Data in LLMs: A Neuron Activation-Based Detection Framework
NA-PDD detects pre-training data in LLMs by comparing which neurons activate for a test text against neurons linked to known training versus non-training texts, and claims large AUC improvements on three benchmarks.
-
Membership Inference Attacks as Privacy Tools: Reliability, Disparity and Ensemble
MIAs expose different members depending on attack method and random seed; the paper quantifies this with coverage/stability and shows ensembling attacks yields stronger, more reliable privacy checks.
- Winter Soldier: Backdooring Language Models at Pre-Training with Indirect Data Poisoning
Discussion (0). Sign in to comment.