An Irreducible Quantum Advantage in Aligning World Models with Reality
Abstract
World models provide digital simulacra of the true world, allowing agents to be trained and tested before costly real-world deployment. At each time step, they receive an action and generate an observation and reward matching the statistics of the true world. In complex environments where present outcomes depend on events far in the past, this requires memory. One might expect that, by increasing memory, we can always build a model accurately enough to align the optimal agent policies of the real and virtual worlds. We show that this is false for classical world models, even when the true world itself is classical. We construct true worlds for which every finite classical model fails along the same possible trajectory: it either loses the ability to distinguish actions when the true world clearly prefers one, or repeatedly assigns the highest expected reward to suboptimal actions. Its expected-reward estimates also retain a nonvanishing average error. In contrast, each such true world admits a quantum world model using a single qutrit that reproduces it exactly: its reward estimates and preferred actions always match those of the true world, ensuring that the optimal policies of the real and virtual worlds remain perfectly aligned.
Related
- Trace-class spectra of irreducible Gaussian quantum Markov semigroups
- Exponential advantage in continuous-variable quantum state learning
- Quantum metrological advantage of high-order squeezed states
- A Framework for Quantum Advantage
- Constant-round quantum advantage in communication complexity for total functions
BibTeX
@misc{lumbreras2026irreducible,
title = {{An Irreducible Quantum Advantage in Aligning World Models with Reality}},
author = {Josep Lumbreras and Hailan Ma and Jayne Thompson and Mile Gu},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2608.19779},
eprint = {2608.19779},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2608.19779},
}
RIS
TY - UNPB
TI - An Irreducible Quantum Advantage in Aligning World Models with Reality
AU - Josep Lumbreras
AU - Hailan Ma
AU - Jayne Thompson
AU - Mile Gu
PY - 2026
DA - 2026/10/02
PB - arXiv
DO - 10.48550/arxiv.2608.19779
UR - https://arxiv.org/abs/2608.19779
AB - World models provide digital simulacra of the true world, allowing agents to be trained and tested before costly real-world deployment. At each time step, they receive an action and generate an observation and reward matching the statistics of the true world. In complex environments where present outcomes depend on events far in the past, this requires memory. One might expect that, by increasing memory, we can always build a model accurately enough to align the optimal agent policies of the real and virtual worlds. We show that this is false for classical world models, even when the true world itself is classical. We construct true worlds for which every finite classical model fails along the same possible trajectory: it either loses the ability to distinguish actions when the true world clearly prefers one, or repeatedly assigns the highest expected reward to suboptimal actions. Its expected-reward estimates also retain a nonvanishing average error. In contrast, each such true world admits a quantum world model using a single qutrit that reproduces it exactly: its reward estimates and preferred actions always match those of the true world, ensuring that the optimal policies of the real and virtual worlds remain perfectly aligned.
ER -