top of page
Andrea Viliotti
30 novTempo di lettura: 9 min
Gaming and Artificial Intelligence. BALROG the New Standard for LLMs and VLMs
BALROG evaluates LLMs/VLMs in complex environments, testing reasoning and planning. It seeks to address agentic and multimodal limitations.
16 visualizzazioni0 commenti
Andrea Viliotti
28 novTempo di lettura: 9 min
GenAI in Banking
GenAI in banking transforms services and risk management through gradual adoption. Governance, ethics, and security are crucial for success.
2 visualizzazioni0 commenti
Andrea Viliotti
28 novTempo di lettura: 10 min
LLMs and Security: MRJ-Agent for a Multi-Round Attack
MRJ-Agent uses risk decomposition and psychological induction for effective multi-round attacks against advanced AI model defenses.
2 visualizzazioni0 commenti
Andrea Viliotti
28 novTempo di lettura: 12 min
BrainBench: Language Models Surpass Neuroscience Experts
BrainBench reveals LLMs predict scientific results with 81.4% accuracy, surpassing human experts at 63.4%.
31 visualizzazioni0 commenti
Andrea Viliotti
17 novTempo di lettura: 14 min
Configurable Foundational Models: A Modular Approach to Building LLMs
Configurable Foundational Models: dynamic LLM modules, more scalable, adaptable, and ideal for efficiency and personalization.
6 visualizzazioni0 commenti
bottom of page