The abstract describes a growing security concern: cellular core-network elements are becoming more exposed to public networks, turning Diameter signaling interfaces into a practical attack surface. This is a shift from the traditional assumption that core networks are isolated and trusted.
The paper investigates whether large language models can be used to assess the exploitability of these interfaces. Rather than relying solely on manual analysis or conventional fuzzing, the authors suggest that LLMs could help reason about protocol-level vulnerabilities in a more automated, field-level manner.
As this is a preprint announcement, the abstract does not include specific findings or validation details. The contribution at this stage is the framing of the problem and the proposal of an LLM-based approach for exploitability analysis in the Diameter ecosystem.