EntropyScan detects backdoored LVLMs by quantifying structural anomalies in visual attention distributions on benign samples via Tsallis entropy and reference-anchored Z-score normalization.
arXiv preprint arXiv:2506.07214 (2025)
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
representative citing papers
ProjLens shows that backdoor parameters in MLLMs are encoded in low-rank subspaces of the projector and that embeddings shift toward the target direction with magnitude linear in input norm, activating only on poisoned samples.
BadVSFM is the first effective backdoor attack on prompt-driven video segmentation foundation models, using a two-stage encoder-decoder strategy to achieve high attack success rates with limited clean performance loss.
Phantasia is a new backdoor attack on VLMs that dynamically aligns malicious outputs with input context to achieve higher stealth and state-of-the-art success rates compared to static-pattern attacks.
citing papers explorer
-
EntropyScan: Towards Model-level Backdoor Detection in LVLMs via Visual Attention Entropy
EntropyScan detects backdoored LVLMs by quantifying structural anomalies in visual attention distributions on benign samples via Tsallis entropy and reference-anchored Z-score normalization.
-
ProjLens: Unveiling the Role of Projectors in Multimodal Model Safety
ProjLens shows that backdoor parameters in MLLMs are encoded in low-rank subspaces of the projector and that embeddings shift toward the target direction with magnitude linear in input norm, activating only on poisoned samples.
-
Backdoor Attacks on Prompt-Driven Video Segmentation Foundation Models
BadVSFM is the first effective backdoor attack on prompt-driven video segmentation foundation models, using a two-stage encoder-decoder strategy to achieve high attack success rates with limited clean performance loss.
-
Phantasia: Context-Adaptive Backdoors in Vision Language Models
Phantasia is a new backdoor attack on VLMs that dynamically aligns malicious outputs with input context to achieve higher stealth and state-of-the-art success rates compared to static-pattern attacks.