machine-learning-safety

Tag

Cards List
#machine-learning-safety

When Certificates Fail: A Unified Safety Framework for Embedded Neural Interface Models

arXiv cs.LG · 5d ago Cached

This paper demonstrates that formal robustness certificates for embedded neural interface models can pass even when task accuracy collapses under adversarial attack, and proposes a unified empirical audit framework to address alignment failures between training objectives and operational user welfare.

0 favorites 0 likes
#machine-learning-safety

Synergistic Simplex: Cooperative Runtime Assurance for Safety-Critical Autonomous Systems

arXiv cs.LG · 2026-05-12 Cached

This paper introduces Synergistic Simplex, a new runtime assurance architecture for autonomous systems that allows safety monitors to use ML outputs while preserving formal safety guarantees. The authors demonstrate its effectiveness in improving performance for obstacle detection in autonomous vehicles.

0 favorites 0 likes
← Back to home

Submit Feedback