dorsal/arxiv
View SchemaLERA: Reinstating Judgment as a Structural Precondition for Execution in Automated Systems
| Authors | Jing, Liu |
|---|---|
| Categories | |
| ArXiv ID | 2601.08880vv1 |
| URL | https://arxiv.org/abs/2601.08880 |
| License | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ |
Abstract
As automated systems increasingly transition from decision support to direct execution, the problem of accountability shifts from decision quality to execution legitimacy. While optimization, execution, and feedback mechanisms are extensively modeled in contemporary AI and control architectures, the structural role of judgment remains undefined. Judgment is typically introduced as an external intervention rather than a native precondition to execution. This work does not propose a new decision-making algorithm or safety heuristic, but identifies a missing structural role in contemporary AI and control architectures. This paper identifies this absence as a missing Judgment Root Node and proposes LERA (Judgment-Governance Architecture) , a structural framework that enforces judgment as a mandatory, non-bypassable prerequisite for execution. LERA is founded on two axioms: (1) execution is not a matter of system capability, but of structural permission, and (2) execution is not the chronological successor of judgment, but its structural consequence. Together, these axioms decouple execution legitimacy from computational capacity and bind it to judgment completion through a governance gate. LERA does not aim to optimize decisions or automate judgment. Instead, it institutionalizes judgment as a first-class architectural component, ensuring that execution authority remains accountable. By reinstating judgment at the execution boundary, LERA establishes a foundational architecture for judgment-governed automation.
{
"annotation_id": "469f3e73-6ba3-4ba4-a285-6aa88919ba83",
"date_created": "2026-02-17T05:53:20.180000Z",
"date_modified": "2026-02-17T05:53:20.180000Z",
"file_hash": "b0c69286d5b7a36994744efa8cd40b6e07fbab3303bcb24e1b37ae285e7b6075",
"private": false,
"record": {
"abstract": "As automated systems increasingly transition from decision support to direct execution, the problem of accountability shifts from decision quality to execution legitimacy. While optimization, execution, and feedback mechanisms are extensively modeled in contemporary AI and control architectures, the structural role of judgment remains undefined. Judgment is typically introduced as an external intervention rather than a native precondition to execution.\n This work does not propose a new decision-making algorithm or safety heuristic, but identifies a missing structural role in contemporary AI and control architectures. This paper identifies this absence as a missing Judgment Root Node and proposes LERA (Judgment-Governance Architecture) , a structural framework that enforces judgment as a mandatory, non-bypassable prerequisite for execution.\n LERA is founded on two axioms: (1) execution is not a matter of system capability, but of structural permission, and (2) execution is not the chronological successor of judgment, but its structural consequence. Together, these axioms decouple execution legitimacy from computational capacity and bind it to judgment completion through a governance gate.\n LERA does not aim to optimize decisions or automate judgment. Instead, it institutionalizes judgment as a first-class architectural component, ensuring that execution authority remains accountable. By reinstating judgment at the execution boundary, LERA establishes a foundational architecture for judgment-governed automation.",
"arxiv_id": "2601.08880",
"authors": [
"Jing",
"Liu"
],
"categories": [
"cs.CY"
],
"license": "http://arxiv.org/licenses/nonexclusive-distrib/1.0/",
"title": "LERA: Reinstating Judgment as a Structural Precondition for Execution in Automated Systems",
"url": "https://arxiv.org/abs/2601.08880",
"version": "v1"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "7dc59891-ab53-4e05-b953-0c1c0848d8e2",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}