As cyberspace continues to expand, identifying the organization or individual behind a website has become increasingly vital in security incident response, phishing website detection, and other cybersecurity subfields. An existing solution for it involves analyzing webpage content and extracting owner names using named entity recognition techniques. However, since these techniques operate on a sentence-by-sentence basis, they struggle to identify the true owner when multiple individual or organizational names appear on a webpage. Moreover, they often perform poorly on non-English websites. To address these limitations, we propose OwnerHunter,1 a novel multilingual framework powered by large language models, which formulates website owner identification as a multilingual document-level information extraction task and utilizes global information from webpages to identify the owner. In OwnerHunter, we first craft prompts that fully leverage the capabilities of large language models to effectively recognize potential owners on webpages in different languages with minimal examples. To enhance the comprehensiveness and accuracy of recognition, we further design a multimodal augmentation strategy, an example pool strategy, and a self-verification strategy. Then, we devise a semantic and string similarity aggregation-based entity disambiguation technique to eliminate ambiguities among multiple potential owners recognized by large language models and a position-based hybrid ranking technique to exactly select the true owner. To evaluate OwnerHunter, we refine the publicly available English dataset ONER and construct the Chinese dataset WOI-cn with 16,036 real websites. Experimental results show that OwnerHunter achieves F1 scores of 0.9505 on ONER and 0.9621 on WOI-cn, setting new state-of-the-art performance on both datasets.¹Codes are available at https://github.com/tuchen9/OwnerHunter

VirtIO devices process guest inputs through virtqueues whose queue roles, buffer directions, scatter–gather layouts, and lengths jointly determine how requests reach device-specific handlers. Existing virtual-device fuzzers primarily mutate low-level PIO, MMIO, and DMA interactions, which often fail to form protocol-consistent VirtIO requests and may collect feedback before deferred processing completes. We present Ouroboros, a protocol-sensitive fuzzing framework that raises the input abstraction to Virtqueue Events and aligns feedback collection with Request Lifecycles. Its protocol-semantics-aware input model separates protocol context from device payload, normalizes shared VirtIO constraints, and materializes requests at the request-consumption boundary. Its protocol-flow-aware execution mechanism tracks each event through synchronous or deferred processing until completion, release, or bounded timeout. We implement Ouroboros on QEMU and integrate it with AFL++ in persistent mode. Across six VirtIO devices and ten 24-h runs per target, Ouroboros increases average final branch coverage from 12.25% to 22.65% and raises the average Valid Input Ratio from 0.24% to 38.47%. It also discovers 10 previously unreported bugs across four QEMU VirtIO device implementations, all of which have been reported to the QEMU maintainers. These results show that aligning fuzzing inputs and feedback with VirtIO request semantics substantially improves exploration of device-specific logic.