# Belief-reality separation lives in routing over a shared value slot in language models

arXiv: https://arxiv.org/abs/2607.11945 · HTML: https://mlsanity.com/paper/2607.11945/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 28 references that are in the map.
- First posted on arXiv 11 Jul 2026.

## Builds on (most cited first, 23 of 23)

- [Discovering Latent Knowledge in Language Models Without Supervision](https://mlsanity.com/paper/2212.03827/) (2022)
- [The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasets](https://mlsanity.com/paper/2310.06824/) (2023)
- [Large Language Models Fail on Trivial Alterations to Theory-of-Mind Tasks](https://mlsanity.com/paper/2302.08399/) (2023)
- [Retrieval Head Mechanistically Explains Long-Context Factuality](https://mlsanity.com/paper/2404.15574/) (2024)
- [Evaluating Large Language Models in Theory of Mind Tasks](https://mlsanity.com/paper/2302.02083/) (2023)
- [Neural Theory-of-Mind? On the Limits of Social Intelligence in Large LMs](https://mlsanity.com/paper/2210.13312/) (2022)
- [Finding Alignments Between Interpretable Causal Variables and Distributed Neural Representations](https://mlsanity.com/paper/2303.02536/) (2023)
- [Understanding Social Reasoning in Language Models with Language Models](https://mlsanity.com/paper/2306.15448/) (2023)
- [FANToM: A Benchmark for Stress-testing Machine Theory of Mind in Interactions](https://mlsanity.com/paper/2310.15421/) (2023)
- [Clever Hans or Neural Theory of Mind? Stress Testing Social Reasoning in Large Language Models](https://mlsanity.com/paper/2305.14763/) (2023)
- [Fine-Tuning Enhances Existing Mechanisms: A Case Study on Entity Tracking](https://mlsanity.com/paper/2402.14811/) (2024)
- [How do Language Models Bind Entities in Context?](https://mlsanity.com/paper/2310.17191/) (2023)
- [Interpretability at Scale: Identifying Causal Mechanisms in Alpaca](https://mlsanity.com/paper/2305.08809/) (2023)
- [Minding Language Models' (Lack of) Theory of Mind: A Plug-and-Play Multi-Character Belief Tracker](https://mlsanity.com/paper/2306.00924/) (2023)
- [Still No Lie Detector for Language Models: Probing Empirical and Conceptual Roadblocks](https://mlsanity.com/paper/2307.00175/) (2023)
- [HI-TOM: A Benchmark for Evaluating Higher-Order Theory of Mind Reasoning in Large Language Models](https://mlsanity.com/paper/2310.16755/) (2023)
- [Language Models use Lookbacks to Track Beliefs](https://mlsanity.com/paper/2505.14685/) (2025)
- [Position: Theory of Mind Benchmarks are Broken for Large Language Models](https://mlsanity.com/paper/2412.19726/) (2024)
- [Standards for Belief Representations in LLMs](https://mlsanity.com/paper/2405.21030/) (2024)
- [Representational Analysis of Binding in Language Models](https://mlsanity.com/paper/2409.05448/) (2024)
- [Perceptions to Beliefs: Exploring Precursory Inferences for Theory of Mind in Large Language Models](https://mlsanity.com/paper/2407.06004/) (2024)
- [How Do Transformers Learn Variable Binding in Symbolic Programs?](https://mlsanity.com/paper/2505.20896/) (2025)
- [Language Statistics and False Belief Reasoning: Evidence from 41 Open-Weight LMs](https://mlsanity.com/paper/2602.16085/) (2026)

## Abstract (arXiv)

Capable language models hold what a character believes apart from what is true: told "Anna believes the cup is blue; in reality it is red," they answer blue about Anna and red about the world. Where in the computation does that separation live? We show it rests on two separable mechanisms at two positions. A generic value slot binds the attributed value. A router at the query position selects which frame, the character's belief or reality, a query reads out. Two routes fill the slot: an asserted belief, whose value the text supplies, binds in directly; a derived belief, whose value must be inferred from what the character could see, arrives by a visibility-gated lookback. A subspace trained on either route steers the other, and only the derived route depends on described visibility. The slot itself carries no belief-reality tag: intervening on it moves a reality readout as strongly as a belief one. The separation lives instead in a dissociated pair of routing subspaces, which flip a query between frames without injecting the donor's value. These results hold across three architectures, on stimuli de-confounded against theory-of-mind-benchmark shortcuts; the behavior itself emerges between 3B and 7B across five model families. This paper develops the single belief-reality axis in depth; a companion paper shows the same slot-and-router format is shared across the other non-actual contexts a sentence can open (counterfactual, fictional, temporal).