| title | Jeva: System One Models from First Principles |
|---|---|
| authors | Qi Li, Xinchao Wang |
| date | 2026-10-13 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2610.18773 |
| affiliation | National University of Singapore |
| upvotes | 180 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf code |
| abstract | System one models enable direct, low-latency decision-making. Yet recent practice of predicting an answer from serialized text does not, by itself, respect the structure of a decision problem. We argue that current designs underemphasize four important structural requirements. (i) null-aware: Models should distinguish answer absence from uncertainty among candidates; (ii) permutation-invariant: Reordering alternatives should not change the decision; (iii) option-scalable: Growing candidate sets should not require redefining the output space; (iv) cacheable: Request-independent computation over recurring alternatives should be reusable across requests. We introduce Jeva, a decision-native architecture that addresses these requirements jointly. A shared encoder independently represents the request and each option, while a compact decision stack refines only the request state against immutable option memory. Symmetric aggregation yields permutation-equivariant probabilities; shared scoring accommodates variable-size candidate sets. A separate coverage estimate distinguishes answer absence from candidate competition, enabling explicit null outcomes. Request-independent option representations and layer projections support reuse across newly assembled sets. These mechanisms unify Choice, Noul, and Score outputs. Across evaluations, Jeva combines strong typed-decision performance with near-perfect null detection and no decision flips under any tested permutations. It scales to thousands of candidates and substantially reduces latency through cross-request reuse without changing decisions. |
cite
@misc{li2026jeva,
title={Jeva: System One Models from First Principles},
author={Qi Li and Xinchao Wang},
year={2026},
eprint={2610.18773},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2610.18773}
}