Timing is Everything: Inferring Temporal Membership in Continually Updated Models

Summary

Researchers present the first systematic study of temporal privacy risk in continually updated deep learning models. They introduce TMIA, a black-box temporal membership inference attack that uses bidirectional LSTMs over model outputs from successive updates to infer whether and when individual samples were used in training. TMIA significantly outperforms existing baselines across three language models and three image classification models.

Topics

• Training data and privacy

Related items

• InfoCoRTeC: A Differentially Private Release for Synthetic Tabular Data, Decodable by Any GeneratorSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)
• InfoProbing Local Conditional Fit for Diffusion Model Membership InferenceSimilar attack · Arxiv (cs.RO + cs.CV security)
• InfoFrom Training Exposure to Membership Evidence: A Controlled Study of OLMo 2Similar attack · Arxiv (cs.CR + cs.CL + cs.LG)
• InfoBeyond public access in LLM pre-training dataSimilar attack · OpenAlex (peer-reviewed AI security)
• InfoCharacterizing Recoverable Expert-Utilization Signatures in Fine-Tuned Mixture-of-Experts LLMsSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)