Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Ji, Deyi, Lu, Junyu, Liu, Xuanyi, Liu, Liqun, Zhang, Hailong, Shu, Peng, Yu, Huan, Jiang, Jie, Chen, Tianru, Zhu, Lanyun
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2605.02200
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866913085416013824
author Ji, Deyi
Lu, Junyu
Liu, Xuanyi
Liu, Liqun
Zhang, Hailong
Shu, Peng
Yu, Huan
Jiang, Jie
Chen, Tianru
Zhu, Lanyun
author_facet Ji, Deyi
Lu, Junyu
Liu, Xuanyi
Liu, Liqun
Zhang, Hailong
Shu, Peng
Yu, Huan
Jiang, Jie
Chen, Tianru
Zhu, Lanyun
contents Online advertising governance faces significant challenges due to the non-stationary nature of regulatory policies, where emerging mandates (e.g., restrictions on education or aesthetic anxiety) create severe label inconsistencies and reasoning ambiguities in historical datasets. In this paper, we propose ARGUS, a policy-adaptive governance system that enables evolving reinforcement through multi-agent adversarial umpiring. ARGUS addresses the sparsity of new policy data by employing a three-stage framework: (1) Policy Seeding for initial perception; (2) Adversarial Label Rectification, which utilizes a ``Prosecutor-Defender-Umpire'' architecture to resolve conflicts between stale labels and new mandates; and (3) Latent Knowledge Discovery, which employs a tripartite dialectical discussion to unearth sophisticated, ``gray-area'' violations. By leveraging RAG-enhanced policy knowledge and Chain-of-Thought synthesis as dynamic rewards for reinforcement learning, ARGUS synchronizes its reasoning pathways with evolving regulations. Extensive experiments on both industrial and public datasets demonstrate that ARGUS significantly outperforms traditional fine-tuning baselines, achieving superior policy-adaptive learning with minimal gold data.
format Preprint
id arxiv_https___arxiv_org_abs_2605_02200
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle ARGUS: Policy-Adaptive Ad Governance via Evolving Reinforcement with Adversarial Umpiring
Ji, Deyi
Lu, Junyu
Liu, Xuanyi
Liu, Liqun
Zhang, Hailong
Shu, Peng
Yu, Huan
Jiang, Jie
Chen, Tianru
Zhu, Lanyun
Computation and Language
Online advertising governance faces significant challenges due to the non-stationary nature of regulatory policies, where emerging mandates (e.g., restrictions on education or aesthetic anxiety) create severe label inconsistencies and reasoning ambiguities in historical datasets. In this paper, we propose ARGUS, a policy-adaptive governance system that enables evolving reinforcement through multi-agent adversarial umpiring. ARGUS addresses the sparsity of new policy data by employing a three-stage framework: (1) Policy Seeding for initial perception; (2) Adversarial Label Rectification, which utilizes a ``Prosecutor-Defender-Umpire'' architecture to resolve conflicts between stale labels and new mandates; and (3) Latent Knowledge Discovery, which employs a tripartite dialectical discussion to unearth sophisticated, ``gray-area'' violations. By leveraging RAG-enhanced policy knowledge and Chain-of-Thought synthesis as dynamic rewards for reinforcement learning, ARGUS synchronizes its reasoning pathways with evolving regulations. Extensive experiments on both industrial and public datasets demonstrate that ARGUS significantly outperforms traditional fine-tuning baselines, achieving superior policy-adaptive learning with minimal gold data.
title ARGUS: Policy-Adaptive Ad Governance via Evolving Reinforcement with Adversarial Umpiring
topic Computation and Language
url https://arxiv.org/abs/2605.02200