Reinforcement Learning from Human Feedback: Whose Culture, Whose Values, Whose Perspectives?
On the map: LLM alignment / RLHF / preference optimisation → Reward modeling and RLHF
Abstract (arXiv)
We argue for the epistemic and ethical advantages of pluralism in Reinforcement Learning from Human Feedback (RLHF) in the context of Large Language Models (LLM). Drawing on social epistemology and pluralist philosophy of science, we suggest ways in which RHLF can be made more responsive to human needs and how we can address challenges along the way. The paper concludes with an agenda for change, i.e. concrete, actionable steps to improve LLM development.
Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Placed in Reward modeling and RLHF, one of 10 directions in LLM alignment / RLHF / preference optimisation. Builds on 16 of its 59 references that are in the map. First posted on arXiv 2 Jul 2024.
Builds on
16 of 59 references.
Closest ancestors · 5
Same direction, newest first
- 2023Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback27 Jul
- 2021Reinforcement Learning with Feedback from Multiple Humans with Diverse Skills16 Nov
- 2020Active Preference-Based Gaussian Process Regression for Reward Learning6 May
- 2019Fine-Tuning Language Models from Human Preferences18 Sep
- 2017Deep reinforcement learning from human preferences12 Jun
Other references · 11
- 2023GPT-4 Technical Reportcited by 18,132
- 2022Training language models to follow instructions with human feedbackcited by 15,712
- 2016Overcoming catastrophic forgetting in neural networkscited by 4,718
- 2022Constitutional AI: Harmlessness from AI Feedbackcited by 2,950
- 2022Discovering Language Model Behaviors with Model-Written Evaluationscited by 805
- 2023Multi-step Jailbreaking Privacy Attacks on ChatGPTcited by 320
- 2022Fine-tuning language models to find agreement among humans with diverse preferencescited by 189
- 2023When Large Language Models Meet Personalization: Perspectives of Challenges and Opportunitiescited by 89
- 2022On the Impossible Safety of Large AI Modelscited by 27
- 2023Towards a Benchmark for Scientific Understanding in Humans and Machinescited by 9
- 2018Targeting Solutions in Bayesian Multi-Objective Optimization: Sequential and Batch Versionscited by 6
Built on by
Citing papers per year: 2025: 3; 2026: 5.
8 papers.
- 2026Long-Tail Knowledge in Large Language Models: Taxonomy, Mechanisms, Interventions and Implicationscited by 2
- 2025Ethics and Persuasion in Reinforcement Learning from Human Feedback: A Procedural Rhetorical Approachcited by 1
- 2025PEPS: Quantum-Inspired Reinforcement Learning for Coherent Reasoning Traces in LLMscited by 0
- 2025A Multifaceted Analysis of Social Biases in Large Language Modelscited by 0
- 2026Preference learning in shades of gray: Interpretable and bias-aware reward modeling for human preferencescited by 0
- 2026Greener Than Humans? Environmental Attitudes in Large Language Modelscited by 0
- 2026Beyond Epistemia: Epistemic Schizologia and Large Language Models as Techno-Semiotic Machinescited by 0
- 2026Controllable Affective Generation via Latent Vector Steeringcited by 0
- 2026Controllable Affective Generation via Latent Vector Steering26 Aug 2026
- 2026Beyond Epistemia: Epistemic Schizologia and Large Language Models as Techno-Semiotic Machines28 Jul 2026
- 2026Greener Than Humans? Environmental Attitudes in Large Language Models1 Jun 2026
- 2026Preference learning in shades of gray: Interpretable and bias-aware reward modeling for human preferences1 Apr 2026
- 2026Long-Tail Knowledge in Large Language Models: Taxonomy, Mechanisms, Interventions and Implications18 Feb 2026
- 2025A Multifaceted Analysis of Social Biases in Large Language Models16 Dec 2025
- 2025PEPS: Quantum-Inspired Reinforcement Learning for Coherent Reasoning Traces in LLMs24 Sep 2025
- 2025Ethics and Persuasion in Reinforcement Learning from Human Feedback: A Procedural Rhetorical Approach14 May 2025
Similar, not cited
Close in wording, no citation either way
- 2024RLHF Deciphered: A Critical Analysis of Reinforcement Learning from Human Feedback for LLMs
- 2024AI Alignment through Reinforcement Learning from Human Feedback? Contradictions and Limitations
- 2023The History and Risks of Reinforcement Learning and Human Feedback
- 2023A Survey of Reinforcement Learning from Human Feedback
- 2024Comparing Rationality Between Large Language Models and Humans: Insights and Open Questions
- 2023Perspectives on the Social Impacts of Reinforcement Learning with Human Feedback
- 2023Reinforcement Learning in the Era of LLMs: What is Essential? What is needed? An RL Perspective on RLHF, Prompting, and Beyond
- 2026Reinforcement Learning from Human Feedback: A Statistical Perspective