CALMA: A Process for Deriving Context-aligned Axes for Language Model Alignment

2025-07-11

AI Summary

Existing AI alignment and evaluation datasets predominantly rely on researcher-defined, Western-centric value axes, neglecting cultural diversity and community heterogeneity in real-world deployment contexts—thereby limiting model responsiveness to actual user needs. This paper introduces CALMA, a novel framework employing an open, participatory, context-grounded methodology. Through cross-cultural community co-design workshops and qualitative analysis, CALMA inductively derives quantifiable alignment dimensions from lived usage experiences. Unlike dominant benchmarks, CALMA eschews implicit value assumptions and systematically uncovers community-prioritized values overlooked by existing evaluations. Piloted across two socioculturally distinct communities, the framework successfully identified novel, context-sensitive alignment dimensions—demonstrating its validity, scalability, and practical utility for pluralistic, transparent AI governance.

Technology Category

• Humans and AI: Learning Human Values and Preferences
• Philosophy and Ethics of AI: Bias, Fairness & Equity
• Natural Language Processing: Ethics — Bias, Fairness, Transparency & Privacy

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Economics, Online Markets and Human Computation: Fairness and ethical considerations in crowd work and in human-in-the-loop AI systems

Abstract

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.

Problem

• Existing alignment datasets overlook diverse real-world user contexts
• Current benchmarks often reflect narrow Western-centric values
• Standard proxies fail to meet needs of varied communities

Innovation

• CALMA derives context-aligned axes for models
• Uses participatory methodology for diverse alignment
• Enables pluralistic context-sensitive alignment pipelines