P-MIA: A Profiled-Based Membership Inference Attack on Cognitive Diagnosis Models
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866914141996843008 |
|---|---|
| author | Hou, Mingliang Wang, Yinuo Guo, Teng Liu, Zitao Dou, Wenzhou Zheng, Jiaqi Luo, Renqiang Tian, Mi Luo, Weiqi |
| author_facet | Hou, Mingliang Wang, Yinuo Guo, Teng Liu, Zitao Dou, Wenzhou Zheng, Jiaqi Luo, Renqiang Tian, Mi Luo, Weiqi |
| contents | Cognitive diagnosis models (CDMs) are pivotal for creating fine-grained learner profiles in modern intelligent education platforms. However, these models are trained on sensitive student data, raising significant privacy concerns. While membership inference attacks (MIA) have been studied in various domains, their application to CDMs remains a critical research gap, leaving their privacy risks unquantified. This paper is the first to systematically investigate MIA against CDMs. We introduce a novel and realistic grey box threat model that exploits the explainability features of these platforms, where a model's internal knowledge state vectors are exposed to users through visualizations such as radar charts. We demonstrate that these vectors can be accurately reverse-engineered from such visualizations, creating a potent attack surface. Based on this threat model, we propose a profile-based MIA (P-MIA) framework that leverages both the model's final prediction probabilities and the exposed internal knowledge state vectors as features. Extensive experiments on three real-world datasets against mainstream CDMs show that our grey-box attack significantly outperforms standard black-box baselines. Furthermore, we showcase the utility of P-MIA as an auditing tool by successfully evaluating the efficacy of machine unlearning techniques and revealing their limitations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_04716 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | P-MIA: A Profiled-Based Membership Inference Attack on Cognitive Diagnosis Models Hou, Mingliang Wang, Yinuo Guo, Teng Liu, Zitao Dou, Wenzhou Zheng, Jiaqi Luo, Renqiang Tian, Mi Luo, Weiqi Cryptography and Security Artificial Intelligence Cognitive diagnosis models (CDMs) are pivotal for creating fine-grained learner profiles in modern intelligent education platforms. However, these models are trained on sensitive student data, raising significant privacy concerns. While membership inference attacks (MIA) have been studied in various domains, their application to CDMs remains a critical research gap, leaving their privacy risks unquantified. This paper is the first to systematically investigate MIA against CDMs. We introduce a novel and realistic grey box threat model that exploits the explainability features of these platforms, where a model's internal knowledge state vectors are exposed to users through visualizations such as radar charts. We demonstrate that these vectors can be accurately reverse-engineered from such visualizations, creating a potent attack surface. Based on this threat model, we propose a profile-based MIA (P-MIA) framework that leverages both the model's final prediction probabilities and the exposed internal knowledge state vectors as features. Extensive experiments on three real-world datasets against mainstream CDMs show that our grey-box attack significantly outperforms standard black-box baselines. Furthermore, we showcase the utility of P-MIA as an auditing tool by successfully evaluating the efficacy of machine unlearning techniques and revealing their limitations. |
| title | P-MIA: A Profiled-Based Membership Inference Attack on Cognitive Diagnosis Models |
| topic | Cryptography and Security Artificial Intelligence |
| url | https://arxiv.org/abs/2511.04716 |