STALA, the Workshop on Security Testing and Assurance for LLMs and Agents, takes place on 22 March 2027 in Seoul, co-located with the Network and Distributed System Security (NDSS) Symposium 2027. Sentinel Intelligence co-founders Lichao Wu and Wenkai Xu are workshop co-chairs, together with Stjepan Picek (University of Zagreb & Radboud University) and Alexandra Dmitrienko (University of Duisburg-Essen). Co-founder Nan Lu is publicity chair.
What it covers
- Discovering vulnerabilities in LLMs and agents.
- Evaluating defenses.
- Testing and coverage.
- Assurance: deciding when a system has been tested enough to trust.
Program
The day includes keynotes, paper sessions, a panel, and paper awards.
- Keynote: Ahmad-Reza Sadeghi, TU Darmstadt
- Keynote: Masashi Sugiyama, RIKEN AIP / The University of Tokyo
- Panel: "When have we tested enough to deploy?"
Call for papers
Accepted papers are published by the Internet Society in the NDSS workshop proceedings. Submissions use the NDSS 2027 templates and are reviewed double-blind. At least one author of each accepted paper must register and present.
- Full papers: up to 8 pages, excluding references.
- Short, position, and experience papers: up to 4 pages, excluding references.
Key dates (all 23:59 AoE):
- Paper submission: 11 December 2026
- Author notification: 1 February 2027
- Camera-ready: 19 February 2027
- Workshop: 22 March 2027
Full submission guidelines are on the workshop website.
Why we're running it
"Is this agent secure enough to deploy?" is the question Sentinel is built to answer. It's still an open research problem, and we think progress on it should happen in the open, with the wider security and machine learning communities.