Anchored Privacy Drifting (APD) replaces privacy-sensitive visual elements with semantically equivalent alternatives while anchoring context, evaluated on the new AdaptShield benchmark with reported gains of 10.4% and 8.5% across four MLLM families.
Dual-priv pruning : Efficient differential private fine-tuning in multimodal large language models, 2025
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Presents SPPE benchmark and ERMA/C2E-S2SER methods for editability assessment and surrogate-to-source recovery in MLLM privacy protection, reporting metric improvements.
citing papers explorer
-
Seeing Without Exposing: Adaptive Privacy Control for Open-World, Context-Hungry MLLMs
Anchored Privacy Drifting (APD) replaces privacy-sensitive visual elements with semantically equivalent alternatives while anchoring context, evaluated on the new AdaptShield benchmark with reported gains of 10.4% and 8.5% across four MLLM families.
-
When Recovery Matters: The Blind Spot of Surrogate Privacy in MLLM Editing
Presents SPPE benchmark and ERMA/C2E-S2SER methods for editability assessment and surrogate-to-source recovery in MLLM privacy protection, reporting metric improvements.