AIBearisharXiv – CS AI · 6h ago7/10
🧠
Analyzing the Narration Gap in LLM-Solver Loops
Researchers identify critical vulnerabilities in LLM-solver hybrid systems where formal verification guarantees break down during the narration phase—converting solver outputs to user-readable answers. Testing five open-source models reveals adversaries can manipulate final responses through prompt injection despite underlying formal correctness, indicating safety-critical applications using AI-assisted reasoning require additional safeguards beyond solver verification.