Three new research papers on arXiv explore distinct applications of AI systems across technical domains.
PRO-V-R1 addresses Register-Transfer Level (RTL) verification, which according to the paper consumes 60-70% of hardware development time. The research introduces a reasoning-enhanced programming agent designed to automate RTL verification tasks using Large Language Models (LLMs), though the abstract notes that prior LLM research has “overwhelmingly centered on RTL g[eneration]” rather than verification.
Orchestrator Multi-Agent Clinical Decision Support System tackles secondary headache diagnosis in primary care settings. According to the paper, “secondary headaches need specialized care and can have devastating consequences if not treated promptly.” The system uses multiple AI agents to identify clinical ‘red flag’ features including “thunderclap onset, meningismus, papilledema, [and] foca[l neurological signs].”
ASTRIDE introduces a security threat modeling platform specifically for AI agent-based systems. The paper states these systems “are becoming increasingly integral to modern software architectures, enabling autonomous decision-making, dynamic task execution, and multimodal interactions through large language models.” However, the research notes these systems “introduce” new security considerations requiring specialized threat modeling approaches.
All three papers are available as preprints on arXiv.