Metraipolis › archive › dc_e75cffe65c47

Quantum-JEPA: Quantum Born Priors for Latent-Action JEPA World Models — arXiv:2610.27921

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.27921
License (as archived)
CC0-1.0
Archived
2026-10-21T07:06:56.006Z
SHA-256 of the archived text
0b9bd8a5b5efbc787c26b9b112e54ebd7186c8a685aa1a5daad9ebcd68a09a36
Stable URL of this page
https://metraipolis.com/documento/dc_e75cffe65c47

How to cite

Quantum-JEPA: Quantum Born Priors for Latent-Action JEPA World Models — arXiv:2610.27921 — arXiv (en), https://arxiv.org/abs/2610.27921 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e75cffe65c47 (Metraipolis archive dc_e75cffe65c47, sha256 0b9bd8a5b5efbc787c26b9b112e54ebd7186c8a685aa1a5daad9ebcd68a09a36)
BibTeX
@misc{dce75cffe65c47,
  title = {Quantum-JEPA: Quantum Born Priors for Latent-Action JEPA World Models — arXiv:2610.27921},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.27921},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e75cffe65c47; sha256 0b9bd8a5b5efbc787c26b9b112e54ebd7186c8a685aa1a5daad9ebcd68a09a36},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.27921. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

A single observation can admit several plausible futures, making uncertainty an essential part of world modelling. However, quantum generative models remain underexplored for this task. We introduce Quantum-JEPA, a latent-action JEPA framework that learns distributions of future representations using a conditional quantum Born prior. Circuit measurements generate binary latent actions, which a shared predictor maps to alternative future representations. We train the quantum prior through the quality and diversity of these predictions using a sampled energy objective. Our model learns distinct continuations on constructed multimodal tasks and captures latent-action correlations that support coordinated multi-agent futures. On D3IL, a benchmark of diverse human demonstrations, our trained models generate multiple obstacle-crossing routes from a common starting point without observed motion. On the controlled tasks, our classical and quantum prior implementations improve prediction validity and distributional fidelity over the evaluated variational baselines. Our quantum priors combine competitive predictive quality with a parameter-efficient sampling structure: they avoid an exponentially sized categorical output layer while supporting correlated latent actions. We further validate finite-shot parameter-shift training, including training with readout noise. Unlike most prior work, which focuses on simulation, we also verify the feasibility of inference with trained priors on a real QPU. These results establish a trainable quantum component for JEPA future prediction and a basis for exploring quantum-enhanced world models at larger scales.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.27921 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-21T07:05:43.455Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).