Jeva：從第一性原理出發的系統一模型

Jeva: System One Models from First Principles

October 13, 2026

作者: Qi Li, Xinchao Wang

cs.AI

摘要

系統一模型能實現直接、低延遲的決策。然而，近期從序列化文字預測答案的做法本身並未遵循決策問題的結構。我們認為，當前設計低估了四項重要的結構性要求。(i) 空值感知：模型應區分答案不存在與候選項之間的不確定性；(ii) 排列不變性：重新排列選項不應改變決策；(iii) 選項可擴展性：候選集合增長不應需要重新定義輸出空間；(iv) 可快取性：對反覆出現的選項進行與請求無關的計算，應能跨請求重複使用。我們提出 Jeva，一個決策原生架構，共同解決這些要求。共享編碼器獨立地表徵請求與每個選項，而緊湊的決策堆疊僅對照不可變的選項記憶來精煉請求狀態。對稱聚合產生排列等變的機率；共享評分可容納可變大小的候選集合。一個獨立的覆蓋率估計區分答案不存在與候選項競爭，從而實現明確的空值結果。與請求無關的選項表徵與層投影支援在新組裝的集合之間重複使用。這些機制統一了 Choice、Noul 與 Score 輸出。在各項評估中，Jeva 兼具強勁的型別化決策效能、近乎完美的空值偵測，以及在任何受測排列下都不會出現決策翻轉。它能擴展至數千個候選項，並透過跨請求重用大幅降低延遲，而不改變決策。

English

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.