Architect-Flow78/axiom-core

Invariant semantic embeddings for LLM safety

1 best points · 1 comments · 1 Hacker News post · 0 GitHub stars · View on GitHub

Hacker News discussions

• We built a tiny invariant layer that preserves meaning when embeddings fail (1 points, 1 comments, by pascalnicolae, 2026-01-28)