arXiv cs.AI· Elinor Poole-Dayan, Jillian Fisher, Jacob Andreas, Mitchell Gordon, Michiel A. Bakker·· 11 小时前AI 评分27
多元价值对齐研究影响力路线图：前沿实验室与开源模型开发者尚无公开采用证据
A Roadmap to Impactful Pluralistic Alignment Research
AI 导读
一项对主要前沿实验室和开源权重开发者的审计发现，多元价值对齐（让 AI 系统代表并服务多元人类价值与视角）尚无公开证据影响其模型训练或评估。作者提出三大障碍并给出研究路线图：现有论证多为规范性或理论性，需测量其对用户与社会的利弊；社区尚未明确多元行为何时必要及具体形态；现有评估与方法未达生产部署要求。
正文 · 原文
Abstract:Much research on responsible AI aims to change the behavior of AI systems that billions of people use. Pluralistic value alignment---the goal of building AI systems that represent and serve diverse human values and perspectives---has emerged as one such active research agenda. Yet, in an audit of the major frontier labs and open-weight developers, we find no public evidence that it has shaped how their models are trained or evaluated. Pluralistic alignment requires adoption to achieve its main goals, so we identify three barriers and lay out a roadmap of research to address each: 1. The justifications for pluralistic alignment so far are mostly normative or theoretical. Studies are needed that measure the benefits and drawbacks of pluralistic AI for users and society. 2. The pluralistic alignment research community has not settled when pluralistic behavior is warranted or what it should look like in practice. Research is needed to specify this behavior concretely enough for developers to operationalize. 3. Existing pluralism evaluations and methods are not ready for deployment. Evaluations and methods are needed that report their effects on other model properties and meet the requirements of production systems. These barriers are not unique to pluralistic alignment, so our roadmap applies to other areas of responsible AI research with similar theories of change. As people hand more decisions to AI systems and AI agents act on their behalf, we hope our roadmap helps this research reach the deployed models making those decisions.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2607.22305 [cs.AI] |
| (or arXiv:2607.22305v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2607.22305 arXiv-issued DOI via DataCite |
Submission history
From: Elinor Poole-Dayan [view email]
[v1]
Fri, 24 Jul 2026 13:46:03 UTC (254 KB)
[v2]
Tue, 3 Nov 2026 18:03:31 UTC (1,492 KB)
来源：arXiv cs.AI · arxiv.org