mlsanityData to 7 Oct 2026

Adaptive Alignment: Dynamic Preference Adjustments via Multi-Objective Reinforcement Learning for Pluralistic AI

arXiv 2410.23630 ↗First posted 31 Oct 2024cs.LG cs.AIcited by 6

Not on the map. Why?

Abstract (arXiv)

Emerging research in Pluralistic Artificial Intelligence (AI) alignment seeks to address how intelligent systems can be designed and deployed in accordance with diverse human needs and values. We contribute to this pursuit with a dynamic approach for aligning AI with diverse and shifting user preferences through Multi Objective Reinforcement Learning (MORL), via post-learning policy selection adjustment. In this paper, we introduce the proposed framework for this approach, outline its anticipated advantages and assumptions, and discuss technical details about the implementation. We also examine the broader implications of adopting a retroactive alignment approach through the sociotechnical systems perspective.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 17 of its 32 references that are in the map. First posted on arXiv 31 Oct 2024.

Builds on

17 of 32 references.

References · 17

  1. 2022Training language models to follow instructions with human feedbackcited by 15,712
  2. 2016Concrete Problems in AI Safetycited by 1,582
  3. 2023Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedbackcited by 743
  4. 2020Artificial Intelligence, Values and Alignmentcited by 443
  5. 2023AI Alignment: A Comprehensive Surveycited by 256
  6. 2021A Practical Guide to Multi-Objective Reinforcement Learning and Planningcited by 240
  7. 2023Rewarded soups: towards Pareto-optimal alignment by interpolating weights fine-tuned on diverse rewardscited by 232
  8. 2024Rewards-in-Context: Multi-objective Alignment of Foundation Models with Dynamic Preference Adjustmentcited by 171
  9. 2017Interactive Learning from Policy-Dependent Human Feedbackcited by 155
  10. 2020Reward-rational (implicit) choice: A unifying formalism for reward learningcited by 119
  11. 2023Value Kaleidoscope: Engaging AI with Pluralistic Human Values, Rights, and Dutiescited by 97
  12. 2018Dynamic Weights in Multi-Objective Deep Reinforcement Learningcited by 96
5 more
  1. 2023AI Alignment and Social Choice: Fundamental Limitations and Policy Implicationscited by 35
  2. 2024Towards Scalable Automated Alignment of LLMs: A Surveycited by 33
  3. 2021Active Learning in Robotics: A Review of Control Principlescited by 26
  4. 2021MORAL: Aligning AI with Human Norms through Multi-Objective Reinforced Active Learningcited by 24
  5. 2023Algorithmic Pluralism: A Structural Approach To Equal Opportunitycited by 13

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 3; 2026: 3.

6 papers.

  1. 2025Direct Alignment with Heterogeneous Preferencescited by 19
  2. 2025Virtual Agent Economiescited by 17
  3. 2025LIVS: A Pluralistic Alignment Dataset for Inclusive Public Spacescited by 8
  4. 2026Overton Pluralistic Reinforcement Learning for Large Language Modelscited by 1
  5. 2026From Sycophantic Consensus to Pluralistic Repair: Why AI Alignment Must Surface Disagreementcited by 1
  6. 2026Meta-ethics and AI: exploring the novel meta-ethical questions in the era of AIcited by 0