AI RESEARCH
SIF: Semantically In-Distribution Fingerprints for Large Vision-Language Models
arXiv CS.CV
•
ArXi:2604.17041v1 Announce Type: new The public accessibility of large vision-language models (LVLMs) raises serious concerns about unauthorized model reuse and intellectual property infringement. Existing ownership verification methods often rely on semantically abnormal queries or out-of-distribution responses as fingerprints, which can be easily detected and removed by adversaries.