OpenLVLM-MIA: A Controlled Benchmark Revealing the Limits of Membership Inference Attacks on Large Vision-Language Models

Documenti analoghi

• Why is the User Interface a Dark Pattern? : Explainable Auto-Detection and its Analysis
  di: Yada, Yuki, et al.
  Pubblicazione: (2023)
• Dark patterns in e-commerce: a dataset and its baseline evaluations
  di: Yada, Yuki, et al.
  Pubblicazione: (2022)
• ImpMIA: Leveraging Implicit Bias for Membership Inference Attack
  di: Golbari, Yuval, et al.
  Pubblicazione: (2025)
• DistractMIA: Black-Box Membership Inference on Vision-Language Models via Semantic Distraction
  di: Tang, Hongyi, et al.
  Pubblicazione: (2026)
• AutoMIA: Improved Baselines for Membership Inference Attack via Agentic Self-Exploration
  di: Liu, Ruhao, et al.
  Pubblicazione: (2026)