A Descriptive and Normative Theory of Human Beliefs in RLHF

• URL: http://arxiv.org/abs/2506.01692v1
• Date: Mon, 02 Jun 2025 13:52:55 GMT
• Title: A Descriptive and Normative Theory of Human Beliefs in RLHF
• Authors: Sylee Dandekar, Shripad Deshmukh, Frank Chiu, W. Bradley Knox, Scott Niekum,
• Abstract summary: We propose that human beliefs about the capabilities of the agent being trained also play a key role in preference generation.<n>We show through synthetic experiments that it is often suboptimal for human preference labelers to assume agent optimality.
• Score: 12.627454162208846
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Human preferences in RLHF are typically modeled as a function of the human's reward function or corresponding optimal state-action values. In this work, we propose that human beliefs about the capabilities of the agent being trained also play a key role in preference generation. We examine two questions related to this hypothesis, one descriptive and one normative, respectively: Do human labelers' beliefs about agent capabilities affect the preferences that they provide? And what is the ideal set of beliefs about an agent -- and resulting preferences -- for humans to have? We propose a new preference model that incorporates human beliefs and provide a normative theory that bounds the error on the final learned policy based on the \textit{mismatch} between the human's beliefs and an idealized set of beliefs. We then confirm via a human study that beliefs about agent capabilities do, in fact, significantly affect preferences and can be influenced through simple interventions. Additionally, we empirically show through synthetic experiments that it is often suboptimal for human preference labelers to assume agent optimality. Collectively, these results theoretically and empirically demonstrate how reducing the mismatch between human beliefs and agent capabilities can lead to more performant RLHF and point toward new best practices for RLHF practitioners.

Related papers

• Belief Attribution as Mental Explanation: The Role of Accuracy, Informativity, and Causality [42.943294683967046]
  We investigate the hypothesis that people prefer to attribute beliefs that are good explanations for the behavior they observe.<n>We develop a computational model that quantifies the explanatory strength of a (natural language) statement about an agent's beliefs.<n>Using this model, we study the role of each factor in how people selectively attribute beliefs to other agents.
  arXiv  Detail & Related papers  (2025-05-26T00:21:38Z)
• Influencing Humans to Conform to Preference Models for RLHF [41.929409024817936]
  A preference model that poorly describes how humans generate preferences risks learning a poor approximation of the human's reward function.<n>We conduct three human studies to asses whether one can influence the expression of real human preferences to more closely conform to a desired preference model.
  arXiv  Detail & Related papers  (2025-01-11T03:12:53Z)
• RLHF from Heterogeneous Feedback via Personalization and Preference Aggregation [24.374185140811115]
  Reinforcement learning from human feedback (RLHF) has been an effective technique for aligning AI systems with human values. In this paper, we focus on addressing the issues due to the inherent heterogeneity in human preferences, as well as their potential strategic behavior in providing feedback. We propose two frameworks to address heterogeneous human feedback in principled ways: personalization-based one and aggregation-based one.
  arXiv  Detail & Related papers  (2024-04-30T23:57:23Z)
• Provable Multi-Party Reinforcement Learning with Diverse Human Feedback [63.830731470186855]
  Reinforcement learning with human feedback (RLHF) is an emerging paradigm to align models with human preferences. We show how traditional RLHF approaches can fail since learning a single reward function cannot capture and balance the preferences of multiple individuals. We incorporate meta-learning to learn multiple preferences and adopt different social welfare functions to aggregate the preferences across multiple parties.
  arXiv  Detail & Related papers  (2024-03-08T03:05:11Z)
• MaxMin-RLHF: Alignment with Diverse Human Preferences [101.57443597426374]
  Reinforcement Learning from Human Feedback (RLHF) aligns language models to human preferences by employing a singular reward model derived from preference data.<n>We learn a mixture of preference distributions via an expectation-maximization algorithm to better represent diverse human preferences.<n>Our algorithm achieves an average improvement of more than 16% in win-rates over conventional RLHF algorithms.
  arXiv  Detail & Related papers  (2024-02-14T03:56:27Z)
• Aligning Large Language Models with Human Preferences through Representation Engineering [41.81020951061438]
  Drawing inspiration from the emerging field of representation engineering (RepE), this study aims to identify relevant representations for high-level human preferences embedded in patterns of activity within an LLM. This novel approach, denoted as Representation Alignment from Human Feedback (RAHF), proves to be effective, computationally efficient, and easy to implement.
  arXiv  Detail & Related papers  (2023-12-26T11:01:36Z)
• Towards Understanding Sycophancy in Language Models [49.352840825419236]
  We investigate the prevalence of sycophancy in models whose finetuning procedure made use of human feedback.<n>We show that five state-of-the-art AI assistants consistently exhibit sycophancy across four varied free-form text-generation tasks.<n>Our results indicate that sycophancy is a general behavior of state-of-the-art AI assistants, likely driven in part by human preference judgments favoring sycophantic responses.
  arXiv  Detail & Related papers  (2023-10-20T14:46:48Z)
• AlignDiff: Aligning Diverse Human Preferences via Behavior-Customisable Diffusion Model [69.12623428463573]
  AlignDiff is a novel framework to quantify human preferences, covering abstractness, and guide diffusion planning. It can accurately match user-customized behaviors and efficiently switch from one to another. We demonstrate its superior performance on preference matching, switching, and covering compared to other baselines.
  arXiv  Detail & Related papers  (2023-10-03T13:53:08Z)
• Provable Benefits of Policy Learning from Human Preferences in Contextual Bandit Problems [82.92678837778358]
  preference-based methods have demonstrated substantial success in empirical applications such as InstructGPT. We show how human bias and uncertainty in feedback modelings can affect the theoretical guarantees of these approaches.
  arXiv  Detail & Related papers  (2023-07-24T17:50:24Z)
• Models of human preference for learning reward functions [80.39289349661364]
  We learn the reward function from human-generated preferences between pairs of trajectory segments. We find this assumption to be flawed and propose modeling human preferences as informed by each segment's regret. Our proposed regret preference model better predicts real human preferences and also learns reward functions from these preferences that lead to policies that are better human-aligned.
  arXiv  Detail & Related papers  (2022-06-05T17:58:02Z)
• Adaptive Agent Architecture for Real-time Human-Agent Teaming [3.284216428330814]
  It is critical that agents infer human intent and adapt their polices for smooth coordination. Most literature in human-agent teaming builds agents referencing a learned human model. We propose a novel adaptive agent architecture in human-model-free setting on a two-player cooperative game.
  arXiv  Detail & Related papers  (2021-03-07T20:08:09Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.