FiMMIA: scaling semantic perturbation-based membership inference across modalities
Anton Emelyanov, Sergei Kudriashov, Alena Fenogenova · 2025
Read FiMMIA: scaling semantic perturbation-based membership inference across modalities by Anton Emelyanov, Sergei Kudriashov, Alena Fenogenova (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2512-02786.txt.
Description
Anton Emelyanov, Sergei Kudriashov, and Alena Fenogenova test whether a multimodal model secretly trained on your data. Membership inference attacks ask if one specific item sat in a model's training set. Many detectors exist for text models, but they falter on multimodal ones.
Extra vision parts and shifts across inputs make multimodal models unstable targets. FiMMIA scales semantic perturbation based inference across modalities, probing how outputs bend under meaning preserving changes. Such audits matter for privacy and copyright, where proving a model trained on protected work has legal weight. Probes that bend sense, not words, test the part of a tool that sees and hears, where leaks are hard to spot with text checks.
Catalog record
| CREATOR | Anton Emelyanov, Sergei Kudriashov, Alena Fenogenova |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 34 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2512-02786