Building Smarter Knowledge Graph Pipeline for Secure Semiconductor Chips
summary
Parses RTL into a knowledge graph that models signals, registers and hierarchies as nodes and data/control/synchronization as edges, scaling from millions of RTL lines to billions of gates. Implements a GraphRAG system that extracts entities from unstructured metadata, computes metrics (influence, distance, signal role) to rank vulnerabilities, and suggests/installs instrumentation for mitigation and ML‑based Trojan detection. A case study on the Ibex processor finds specific desynchronization and spoofing exploit locations, demonstrating the pipeline is practical and effective.
AI-written summary · about
chapters (12)
- 0:00Introduction to the challenge
- 3:34Threat landscape overview
- 5:07Vulnerability trends
- 7:11Impact on critical infrastructure
- 9:49Limitations of formal methods
- 10:26Introducing knowledge graphs
- 10:50Knowledge graph methodology
- 14:02Graph elements and structure
- 14:57Hierarchical graph construction
- 21:52Quantitative vulnerability analysis
- 23:44Applications and instrumentation
- 25:44Case study: Ibex processor
related talks
34mIt's Dangerous To SLSA Alone Out There! Take This Artifact Knowledge Graph!
35mComplex Querying of Cyber Knowledge Graph for Robust Cyber Defense
45mIntegrated Circuit Offensive Security
32mBuilding Intelligent Automatons with Semantic Reasoning
29mKnowledge Graphs in Action: Identifying Sanctions Circumvention in a Complex Global Landscape
27mHow Trustworthy is Your Knowledge Graph?
45mConnecting the dots to go from tabular security incident data to behavioral graph understanding
30mCybersecurity Risk Assessment Using LLM Agents and Graph Data Science