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

arXiv: https://arxiv.org/abs/2507.09060 · HTML: https://mlsanity.com/paper/2507.09060/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in LLM preference optimization methods, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 19 of its 32 references that are in the map.
- First posted on arXiv 11 Jul 2025.

## Placed in

- LLM alignment / RLHF / preference optimisation / [LLM preference optimization methods](https://mlsanity.com/alignment/llm-preference-optimization-methods/)

## Builds on (most cited first, 19 of 19)

- [Constitutional AI: Harmlessness from AI Feedback](https://mlsanity.com/paper/2212.08073/) (2022)
- [Improving alignment of dialogue agents via targeted human judgements](https://mlsanity.com/paper/2209.14375/) (2022)
- [FUDGE: Controlled Text Generation With Future Discriminators](https://mlsanity.com/paper/2104.05218/) (2021)
- [Large Language Model Alignment: A Survey](https://mlsanity.com/paper/2309.15025/) (2023)
- [Fine-tuning language models to find agreement among humans with diverse preferences](https://mlsanity.com/paper/2211.15006/) (2022)
- [A Roadmap to Pluralistic Alignment](https://mlsanity.com/paper/2402.05070/) (2024)
- [Process for Adapting Language Models to Society (PALMS) with Values-Targeted Datasets](https://mlsanity.com/paper/2106.10328/) (2021)
- [Arithmetic Control of LLMs for Diverse User Preferences: Directional Preference Alignment with Multi-Objective Rewards](https://mlsanity.com/paper/2402.18571/) (2024)
- [On Releasing Annotator-Level Labels and Information in Datasets](https://mlsanity.com/paper/2110.05699/) (2021)
- [Collective Constitutional AI: Aligning a Language Model with Public Input](https://mlsanity.com/paper/2406.07814/) (2024)
- [SteerLM: Attribute Conditioned SFT as an (User-Steerable) Alternative to RLHF](https://mlsanity.com/paper/2310.05344/) (2023)
- [Aligning Large Language Models through Synthetic Feedback](https://mlsanity.com/paper/2305.13735/) (2023)
- [Binary Classifier Optimization for Large Language Model Alignment](https://mlsanity.com/paper/2404.04656/) (2024)
- [Generative Social Choice](https://mlsanity.com/paper/2309.01291/) (2023)
- [Second Thoughts are Best: Learning to Re-Align With Human Values from Text Edits](https://mlsanity.com/paper/2301.00355/) (2023)
- [Indian-BhED: A Dataset for Measuring India-Centric Biases in Large Language Models](https://mlsanity.com/paper/2309.08573/) (2023)
- [Wikibench: Community-Driven Data Curation for AI Evaluation on Wikipedia](https://mlsanity.com/paper/2402.14147/) (2024)
- [Pitfalls of Evidence-Based AI Policy](https://mlsanity.com/paper/2502.09618/) (2025)
- [LocalValueBench: A Collaboratively Built and Extensible Benchmark for Evaluating Localized Value Alignment and Ethical Safety in Large Language Models](https://mlsanity.com/paper/2408.01460/) (2024)

## Abstract (arXiv)

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.