The Severance Problem: LLMs Lack User Representation Beyond the Prompt

Jul 17, 2026

A new preprint introduces the 'Severance Problem,' highlighting that large language models (LLMs) do not explicitly represent what they do not know about users beyond the immediate prompt. The authors propose the 'Severance Schema,' a method that structures the model's ignorance about the user along several dimensions. Empirical tests across five model families show that this schema reduces undesirable behaviors such as sycophancy, harmful advice, and hallucination, and encourages models to ask clarifying questions when user information is missing.

Why it matters: This work proposes a practical method to address a core limitation in personal AI assistants, potentially improving their safety and reliability.

Full story at: arXiv Computation and Language ↗