# A Method for Learning Value Systems in Generative AI

arXiv: https://arxiv.org/abs/2607.16903 · HTML: https://mlsanity.com/paper/2607.16903/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 38 references that are in the map.
- First posted on arXiv 18 Jul 2026.

## Builds on (most cited first, 11 of 11)

- [Convex optimization](https://mlsanity.com/paper/2106.01946/) (2021)
- [BeaverTails: Towards Improved Safety Alignment of LLM via a Human-Preference Dataset](https://mlsanity.com/paper/2307.04657/) (2023)
- [Personalized Soups: Personalized Large Language Model Alignment via Post-hoc Parameter Merging](https://mlsanity.com/paper/2310.11564/) (2023)
- [Inverse Reward Design](https://mlsanity.com/paper/1711.02827/) (2017)
- [Helping or Herding? Reward Model Ensembles Mitigate but do not Eliminate Reward Hacking](https://mlsanity.com/paper/2312.09244/) (2023)
- [SteerLM: Attribute Conditioned SFT as an (User-Steerable) Alternative to RLHF](https://mlsanity.com/paper/2310.05344/) (2023)
- [Controllable Preference Optimization: Toward Controllable Multi-Objective Alignment](https://mlsanity.com/paper/2402.19085/) (2024)
- [On the Fenchel Duality between Strong Convexity and Lipschitz Continuous Gradient](https://mlsanity.com/paper/1803.06573/) (2018)
- [Align Anything: Training All-Modality Models to Follow Instructions with Language Feedback](https://mlsanity.com/paper/2412.15838/) (2024)
- [LoRe: Personalizing LLMs via Low-Rank Reward Modeling](https://mlsanity.com/paper/2504.14439/) (2025)
- [An Optimal Multistage Stochastic Gradient Method for Minimax Problems](https://mlsanity.com/paper/2002.05683/) (2020)

## Abstract (arXiv)

Value-aware AI systems require explicit computational representations of human values (groundings) and their aggregation into value systems in order to align their decisions with ours. As such representations are difficult to elicit, value learning seeks to infer them by observing human behaviour. This work addresses the lack of grounded value learning methods in generative AI: existing approaches typically replicate human preferences without awareness of the multidimensional structure of value alignment, or lack principled value system elicitation methods. To address these gaps, we adapt a previously validated value system learning method to the generative AI setting, which, based on pairwise prompt-response preference data, simultaneously learns: i) an implementation of a grounding for a set of values given by a multi-objective reward model, and ii) a value system representation in the form of a weighted linear scalarization of the previous grounding model. To ensure that the learned value systems are based on coherent value representations, our algorithm dynamically prioritizes the grounding learning process. We evaluate the method against baselines and a contemporary method on prompt-response preference datasets. Results show competitive performance and minimal trade-offs against the baselines, while improving explainability.