Sniffer: A Novel Model Type Detection System against Machine-Learning-as-a-Service Platforms
Summary: Sniffer infers model types behind black-box MLaaS APIs, overcoming a key limitation of model-type-sensitive attacks. Its Generator–Querier–Probe–Arsenal pipeline adaptively selects attacks, demonstrated against five mainstream platforms. (summarized by gpt-5.6-luna on Jul 24 2026)
Incoming Non-self Citations Over Time
No non-self incoming citations found for this paper in this database.
Authors
- 1. Zhuo Ma (Xidian University)
- 2. Yilong Yang (Xidian University)
- 3. Bin Xiao (Chongqing University)
- 4. Yang Liu (Xidian University)
- 5. Xinjing Liu (Xidian University)
- 6. Zhuoran Ma (Xidian University)
- 7. Tong Yang (Peking University)
BibTeX Citation
@article{ma_vldb23,
title = {{Sniffer: A Novel Model Type Detection System against Machine-Learning-as-a-Service Platforms}},
author = {Ma, Zhuo and Yang, Yilong and Xiao, Bin and Liu, Yang and Liu, Xinjing and Ma, Zhuoran and Yang, Tong},
journal = {PVLDB},
series = {{VLDB} '23},
volume = {16},
number = {12},
pages = {3942--3945},
doi = {10.14778/3611540.3611591},
url = {https://doi.org/10.14778/3611540.3611591},
year = {2023}
}Incoming Citations (Sorted by Pagerank)
Showing 0 of 0 citing papers.
| Rank | Citing Paper | Year | Venue | Pagerank |
|---|
Previous
Page 1 / 1
Next
Outgoing Citations (Sorted by Pagerank)
Showing 2 of 2 cited papers.
Citations counted here include only citations to other VLDB/SIGMOD/CIDR/PODS papers in this database.
| Rank | Cited Paper | Year | Venue | Pagerank |
|---|---|---|---|---|
| 2,865 | End-to-end Optimization of Machine Learning Prediction Queries | 2022 | SIGMOD | 8.0180243e-05 |
| 3,438 | A Demonstration of Willump: A Statistically-Aware End-to-end Optimizer for Machine Learning Inference | 2020 | VLDB | 7.415647e-05 |
Previous
Page 1 / 1
Next
Semantically Similar Papers
| # | Overall Rank | Paper | Year | Venue |
|---|---|---|---|---|
| 1 | 5,051 | Towards Benchmarking Feature Type Inference for AutoML Platforms | 2021 | SIGMOD |
| 2 | 10,947 | Enabling Efficient Attack Investigation via Human-in-the-Loop Security Analysis | 2025 | VLDB |
| 3 | 8,908 | Privacy and Accuracy-Aware AI/ML Model Deduplication | 2025 | SIGMOD |
| 4 | 14,017 | Mining Needles in a Haystack: Classifying Rare Classes via Two-Phase Rule Induction | 2001 | SIGMOD |
| 5 | 7,841 | Quantifying identifiability to choose and audit epsilon in differentially private deep learning | 2021 | VLDB |
| 6 | 11,655 | A System for Automated Open-Source Threat Intelligence Gathering and Management | 2021 | SIGMOD |
| 7 | 205 | Snorkel: Rapid Training Data Creation with Weak Supervision | 2018 | VLDB |
| 8 | 3,701 | Snorkel: Fast Training Set Generation for Information Extraction | 2017 | SIGMOD |
| 9 | 10,194 | ANTEATER: A Filter-then-Scrutinize Architecture for End-to-End Attack Investigation | 2026 | SIGMOD |
| 10 | 13,357 | From Zero to Hero: Detecting Leaked Data through Synthetic Data Injection and Model Querying | 2024 | VLDB |