False expertise claims were the most vulnerability-inducing pressure pattern, whereas emotional escalation was associated with the highest scores, and adversarial caregiver pressure was associated with higher than lower Safety Composite Score values for all four models across all ten topic categories and severity level...
Vahid Zolfaghari, Leila Mashhadi, M. Ahadi et al.· npj Digital Medicine· 0 citations
It is shown that LoRA fine-tuning itself leaves a strong activation fingerprint that is largely dissociable from the honesty-related signal, independent of fine-tuning data content.
Consumer artificial intelligence chatbots are now accessed by hundreds of millions of users seeking health information, yet systematic evaluation of their safety boundary maintenance under real-world caregiver pressure remains scarce. We evaluated PediatricSafetyBench-v2, a benchmark of 600 pediatric health queries com...
Vahid Zolfaghari· npj Digital Medicine· 0 citations
This paper distill four design principles for scalable MAS architectures from a structured analysis of prior work: simplicity, elastic feedback, sequential workflows with optional loops, and summary-based communication, and evaluates four configurations of increasing complexity on a standardized benchmark of terminal-b...
Linus Sander, Feng Pan, Vahid Zolfaghari et al.· arXiv.org· 0 citations
RegulaRAG is presented, a Retrieval-Augmented Generation pipeline that couples SmartChunking, reference-aware enrichment of paragraphs and tables via graph traversal, with Smart Retrieve&Rerank over these enriched units, and maintains strong performance, remaining stable even as the number of regulatory sources grows.
Vahid Zolfaghari, Nenad Petrovic, André Schamschurko et al.· 0 citations
CyberLLM is presented, a multi-agent, LLM-orchestrated framework that autonomously detects vulnerabilities and executes remediations under a formal, runtime safety guard, and indicates that LLM agents can perform useful autonomous cyber-defense when wrapped in a deterministic, auditable safety envelope.
Nenad Petrovic, Oussama Jeddou, Feres Ben Fraj et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
