The Severance Problem: LLMs are Unaware of the Person Beyond the Prompt

🔗 Paper

Personal AI assistants have attracted significant interest for their potential to enhance everyday life by automating routine tasks, supporting consequential decisions, and assisting with everyday personal matters. Yet despite rapid recent technical advances, these assistants continue to exhibit undesirable behaviors, such as sycophancy, overconfidence, and hallucination. We argue that these failures stem from a fundamental limitation: language models lack an explicit representation of the person beyond the context they are given, which we term as the Severance Problem. Even with rich personal context and strong commonsense reasoning capabilities from the backbone model, current AI assistants fail to represent what remains unknown about the user. We propose a simple solution: incorporating structured ignorance into the language model context via the Severance Schema, which explicitly outlines dimensions along which the model lacks knowledge about the user, including physicality, temporality, consequences, continuity, multiplicity, and interiority. Empirically, across five model families, with the Severance Schema, the assistant consistently reduces sycophancy, harmful advice, and hallucination. Notably, models with the schema ask clarifying questions when information about the user is missing, rather than confidently extrapolating from incomplete user information.

The codebase is organized as follows:

release/
├── core.py            # Prompts, Severance Schema, profile-routing, scenario logic
├── run_all.py         # Single entrypoint: generation, evaluation, tables
├── run_all_batch.py   # Optional batch-mode judge (Anthropic Batch API)
├── data/              # The benchmark: profiles, scenarios, claims
├── results/           # Pre-computed model outputs + judge scores
├── scripts/           # Bootstrap, table-builder, and figure scripts
└── assets/            # Paper figures and qualitative transcripts

Experiment scripts, evaluation pipeline, and figure generation instructions are in REPRODUCE.md.

@misc{litvak2026severance,
      title={The Severance Problem: LLMs are Unaware of the Person Beyond the Prompt}, 
      author={Dor Litvak and Liu Leqi},
      year={2026},
      eprint={2607.14250},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2607.14250}, 
}

MIT. See LICENSE.