Inverting the Hidden: Unveiling Multimodal Privacy Leakage in Collaborative LVLM Inference

arXiv:2608.01020 · cs.CR · Submitted 2026-08-02 · Read on arXiv

Shuaifan Jin, Zhibo Wang, Qiyuan Wang, Yiting Han, Yajie Zhou, Yuanfan Zhang, Jiahui Hu, Xiaoyi Pang

cs.CR

Submitted: 2026-08-02

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Transformer — 27× in this paper · explained in 2 episodes
• Consistency — 12× in this paper · explained in 3 episodes
• Accuracy — 10× in this paper · explained in 1 episode
• Decoder — 6× in this paper · explained in 2 episodes
• Vision-language models — 6× in this paper · explained in 2 episodes
• Computer Vision and Pattern Recognition — 5× in this paper
• Edge Devices — 5× in this paper · explained in 2 episodes
• Embeddings — 5× in this paper · explained in 2 episodes
• Ablation — 4× in this paper · explained in 1 episode
• Benchmark — 3× in this paper · explained in 2 episodes
• Transformers — 3× in this paper · explained in 2 episodes
• Cross-modal Alignment — 2× in this paper · explained in 2 episodes
• Fidelity — 2× in this paper · explained in 2 episodes
• Fine-Tuning — 2× in this paper · explained in 5 episodes
• Multimodal Inputs — 2× in this paper · explained in 2 episodes
• Verification — 2× in this paper · explained in 2 episodes
• Ablation Study — 1× in this paper · explained in 3 episodes
• Abstraction — 1× in this paper · explained in 1 episode
• Clarity — 1× in this paper · explained in 3 episodes
• Communication Overhead — 1× in this paper · explained in 2 episodes
• Decode — 1× in this paper · explained in 2 episodes
• Degeneracy — 1× in this paper · explained in 4 episodes
• Exact Match Rate — 1× in this paper
• IEEE Symposium on Security and Privacy — 1× in this paper
• Information Leakage — 1× in this paper · explained in 2 episodes
• Instruction Tuning — 1× in this paper · explained in 3 episodes
• Localization — 1× in this paper · explained in 2 episodes
• Parameterization — 1× in this paper · explained in 2 episodes
• Recoverability-Aware Symmetric Reconstruction — 1× in this paper
• Semantic Similarity — 1× in this paper · explained in 2 episodes
• Transformer Model — 1× in this paper · explained in 2 episodes
• Transformer Models — 1× in this paper · explained in 3 episodes

Sources

• Language Models are Injective and Hence Invertible
• Rep2Text: Decoding Full Text from a Single LLM Token Representation
• Understanding Transformer-Based Vision Models via Modular Feature Inversion
• Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond
• Qwen2.5-VL Technical Report
• Qwen3-VL Technical Report
• SplitLLM: Collaborative Inference of LLMs for Model Placement and Throughput Optimization
• What do we learn from inverting CLIP models?
• Do Vision-Language Models Leak What They Learn? Adaptive Token-Weighted Model Inversion Attacks
• FEVER: a large-scale dataset for Fact Extraction and VERification

Related papers

• SoK: AI-Augmented Binary Reversing
• Relaxed Sender Anonymity for CBDC Interbank Settlement: A Zero-Knowledge Approach on Permissioned EVM
• Calibration-Family Overfit: Why Trusted Sabotage Monitors Don't Transfer Across Lineages
• Estimating the Social Cost of Corporate Data Breaches
• Efficient Fuzzy PSI under One-Sided Assumptions
• Sealing the Audit-Runtime Gap for LLM Skills