Publications

Preprints

Published in arXiv, 2026

An agentic system that generates and validates TLA+ specifications, finds deep bugs through model checking, and reproduces them in system code.

Published in ICLR 2026, 2026

A benchmark suite for evaluating AI capabilities on formally modeling complex real-world systems.

Published in Journal of Software, 2025

This paper provides a comprehensive survey of research progress in distributed system model checking.