Research map / Learning theory and algorithms / Prediction and decision losses
Reference Recommendation Based Membership Inference Attack Against Hybrid-Based Recommender Systems
Xiaoxiao Chi, Xuyun Zhang, Yan Wang, Hongsheng Hu, Wanchun Dou
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use reference recommendation (attribute-only query) in relative metric instead of shadow-model attack | MIA on DropoutNet · ML-1M (target DropoutNet) | ASR: 0.5139 -> 0.9340 | Improved |
| use reference recommendation relative metric instead of ST-MIA | MIA on Heater · ML-100K (target Heater) | TPR@1%FPR: 3.18% -> 56.05% | Improved |
| replace shadow-training attacks with shadow-free metric | MIA · Attack time | time (s): 973.3 s -> 10.4 s | Improved |
| increase number of recommendations n from 10 to 100 (ablation) | reference-recommendation MIA · ML-100K (target DropoutNet) | ASR: below 0.9 -> above 0.9 | Improved |
| increase item feature vector length l from 10 to 100 (ablation) | reference-recommendation MIA · all four settings | ASR | No clear effect |
| add Gaussian-noise DP on member histories (epsilon 0.1) (ablation) | target DropoutNet · ML-100K (target DropoutNet) | ASR of attack: 0.9098 -> 0.5101 | Hurt |
Closest papers by meaning
- Membership Inference Attacks against Fine-tuned Large Language Models via Self-prompt Calibration (NeurIPS 2024 · similarity 0.84)
- Causal Evaluation of Membership Inference Attacks (NeurIPS 2026 · similarity 0.84)
- Exploring the limits of strong membership inference attacks on large language models (NeurIPS 2025 · similarity 0.83)
- Trust-GRS: A Trustworthy Training Framework for Graph Neural Network Based Recommender Systems Against Shilling Attacks (AAAI 2025 · similarity 0.82)
- Practical Bayes-Optimal Membership Inference Attacks (NeurIPS 2025 · similarity 0.82)
- Would You Like Your Data to Be Trained? A User Controllable Recommendation Framework (AAAI 2024 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
