Metraipolis › archive › dc_a153b46f2ea8

DES-JEPA: Learning a Shared Vocabulary for High-Dimensional Discrete Event Sequences — arXiv:2610.38445

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.38445
License (as archived)
CC0-1.0
Archived
2026-10-31T06:08:56.472Z
SHA-256 of the archived text
70941dc358fe0243b6c60e055eecd2fe567248407fe44c99d6053bb155eda46d
Stable URL of this page
https://metraipolis.com/documento/dc_a153b46f2ea8

How to cite

DES-JEPA: Learning a Shared Vocabulary for High-Dimensional Discrete Event Sequences — arXiv:2610.38445 — arXiv (en), https://arxiv.org/abs/2610.38445 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a153b46f2ea8 (Metraipolis archive dc_a153b46f2ea8, sha256 70941dc358fe0243b6c60e055eecd2fe567248407fe44c99d6053bb155eda46d)
BibTeX
@misc{dca153b46f2ea8,
  title = {DES-JEPA: Learning a Shared Vocabulary for High-Dimensional Discrete Event Sequences — arXiv:2610.38445},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.38445},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a153b46f2ea8; sha256 70941dc358fe0243b6c60e055eecd2fe567248407fe44c99d6053bb155eda46d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.38445. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Temporal foundation models such as Chronos and TimesFM generalize across domains through a shared numerical observation space. In contrast, discrete event sequences (DESs), including operational event logs or patient clinical-event trajectories, have domain-specific event vocabularies whose labels lack a consistent correspondence across domains. Beyond this lack of consistency, the number of distinct events in real-world settings is high enough to challenge both classical temporal point process modeling and current in-context foundation inference models such as FIM-PP. Critically, current foundation models for DESs are limited to a small, fixed number of event types, restricting their applicability to realistic, high-dimensional event streams. To address this, we propose two complementary modules: (i) DES-JEPA, a joint-embedding predictive architecture that learns a shared discrete vocabulary from synthetic stochastic processes with unaligned event vocabularies; and (ii) DES-PFN, a discrete event sequence prior-fitted network that reuses the learned vocabulary for in-context next-event prediction. Without task-specific fine-tuning, this framework achieves competitive zero-shot performance on eight unseen real-world datasets and scales up to 700 distinct event types.

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/2610.38445 — 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 2026-10-31T06:07:08.477Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).