pres_security_benchmarking_llm/slides.md

theme	title	titleTemplate	author	info	keywords	mdc	hideInToc	addons	python	presenter	browserExporter	download	exportFilename	twoslash	lineNumbers	monaco	selectable	record	contextMenu	wakeLock	overviewSnapshots	colorSchema	routerMode	aspectRatio	canvasWidth	css	unocss	defaults	drawings	htmlAttrs	layout	transition	background
seriph	Security Benchmarking for Large Language Models	%s - Security Benchmarking for AI	Rossi Stefano	## Methodologies and Applications	Security, Benchmarking, Large Language Models, LLM, Red Teaming	true	false	slidev-addon-rabbit slidev-addon-python-runner	installs prelude loadPackagesFromImports suppressDeprecationWarnings alwaysReload loadPyodideOptions deepeval GREETING_FROM_PRELUDE = "Hello, Slidev!" true true false	true	dev	true	security-benchmarking-for-llms	false	true	false	false	dev	dev	true	false	dark	history	16/9	980	unocss	configFile ./uno.config.ts	layout default	enabled persist presenterOnly syncAll true false false true	dir lang ltr en	center	slide-left	none

Security Benchmarking for Large Language Models

Methodologies and Applications

Understanding, Evaluating, and Mitigating LLM Security Vulnerabilities

Rossi Stefano

14 March, 2025

---

src: ./pages/intro.md transition: fade

---

src: ./pages/risks-and-vulnerabilities.md transition: slide-up

---

src: ./pages/understanding-llm-vulns.md transition: slide-right

---

src: ./pages/red-teaming-methodologies.md transition: zoom-in

---

src: ./pages/advanced-att-techniques.md transition: slide-left

---

src: ./pages/major-bench-secu.md transition: fade-out

---

src: ./pages/deepeval-example.md transition: zoom-out

---

src: ./pages/best-practices.md transition: slide-up

---

src: ./pages/end.md transition: fade