In the frame of the Seminars from the Industry module, AI4CI Paris is hosting talk by Lorenzo Maggi, Senior Research Scientist at Nvidia. When: April 17, 2026 from 10h00 to 12h00Where: Room 33.1.27C at CNAM, 2 rue Conté, 75003 Paris, France Speaker:
Professors' latest publications
- ALARM: Adaptive Layer-Aware Resource Management for Power-Efficient vRANs
- PQTrust-Agent: Policy-Compiled Post-Quantum Trust Contracts for Web-of-Agents Communications
- Evaluating Indirect Prompt Injection in RAG Systems: Defense-in-Depth, Retrieval Effects, and Security-Utility Trade-offs
- Measuring the Operational Value of Identity-Driven Micro-Segmentation in Hybrid Environments: A Reproducible Comparison with ACL-Based Segmentation
- Building a Community-Grounded Spell Checker for the Kabyle Language Using NLP Powered by AI
- From Attack Scenario to Measurable Detection Improvement: A Knowledge-Driven ATT&CK/D3FEND Framework for Instrumented Purple Teaming
- CLIN-FL: Cost-efficient Load-balancing for In-Network Federated Learning with non-IID data
- CIGMA: An Evidence-Grounded Cryptographic Inventory Graph for Reproducible Post-Quantum Migration Prioritization
- Context vs. Compute: Self-Hosted Detection Rule Generation with Knowledge-Graph-Augmented Language Models
- AgentFlowShield: Defending LLM Agents Against Metadata Side Channels
- Occamy: Latency-Aware Scheduling with Controlled Degradation for Mixed-Criticality AI at the Edge
- The Life of a Token: from Words to Bits on the Wire
- GovSecLLM++: A Compliance-Aware Benchmark for Security Testing and Governance Evidence in LLM-Based Applications
- Knowledge-Graph-Constrained LLM Generation of SIEM Detection Rules
- Benchmarking Detection Engineering Improvements through Instrumented Adversary Emulation
