CALMA: A Process for Deriving Context-aligned Axes for Language Model Alignment
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866912484010491904 |
|---|---|
| author | Soni, Prajna Raman, Deepika Hadfield-Menell, Dylan |
| author_facet | Soni, Prajna Raman, Deepika Hadfield-Menell, Dylan |
| contents | Datasets play a central role in AI governance by enabling both evaluation (measuring capabilities) and alignment (enforcing values) along axes such as helpfulness, harmlessness, toxicity, quality, and more. However, most alignment and evaluation datasets depend on researcher-defined or developer-defined axes curated from non-representative samples. As a result, developers typically benchmark models against broad (often Western-centric) values that overlook the varied contexts of their real-world deployment. Consequently, models trained on such proxies can fail to meet the needs and expectations of diverse user communities within these deployment contexts. To bridge this gap, we introduce CALMA (Context-aligned Axes for Language Model Alignment), a grounded, participatory methodology for eliciting context-relevant axes for evaluation and alignment. In a pilot with two distinct communities, CALMA surfaced novel priorities that are absent from standard benchmarks. Our findings demonstrate the value of evaluation practices based on open-ended and use-case-driven processes. Our work advances the development of pluralistic, transparent, and context-sensitive alignment pipelines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_09060 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | CALMA: A Process for Deriving Context-aligned Axes for Language Model Alignment Soni, Prajna Raman, Deepika Hadfield-Menell, Dylan Computers and Society Datasets play a central role in AI governance by enabling both evaluation (measuring capabilities) and alignment (enforcing values) along axes such as helpfulness, harmlessness, toxicity, quality, and more. However, most alignment and evaluation datasets depend on researcher-defined or developer-defined axes curated from non-representative samples. As a result, developers typically benchmark models against broad (often Western-centric) values that overlook the varied contexts of their real-world deployment. Consequently, models trained on such proxies can fail to meet the needs and expectations of diverse user communities within these deployment contexts. To bridge this gap, we introduce CALMA (Context-aligned Axes for Language Model Alignment), a grounded, participatory methodology for eliciting context-relevant axes for evaluation and alignment. In a pilot with two distinct communities, CALMA surfaced novel priorities that are absent from standard benchmarks. Our findings demonstrate the value of evaluation practices based on open-ended and use-case-driven processes. Our work advances the development of pluralistic, transparent, and context-sensitive alignment pipelines. |
| title | CALMA: A Process for Deriving Context-aligned Axes for Language Model Alignment |
| topic | Computers and Society |
| url | https://arxiv.org/abs/2507.09060 |