When Preferences Fail to Become Incentives: A Utility-Behavior Gap in Large Language Models

• URL: http://arxiv.org/abs/2606.22974v2
• Date: Tue, 23 Jun 2026 05:42:48 GMT
• Title: When Preferences Fail to Become Incentives: A Utility-Behavior Gap in Large Language Models
• Abstract summary: We investigate whether preferences serve as motivations for LLM behavior in realistic scenarios.<n>We show that LLMs can be motivated via direct exhortation and other explicit cues to modulate their output quality.<n>We conclude that the existence of coherent preferences should not be taken as evidence that those preferences have incentive value for the models or affect their behavior in other contexts.
• Score: 1.21057012246126
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent work on preference elicitation in large language models (LLMs) has demonstrated that, when given a series of choices between two outcomes, LLMs reveal a coherent, model-specific utility structure. Notably, this structure often includes preferences that the models' trainers did not intend, such as valuing people of some nationalities above others, raising the possibility that LLMs might be forming emergent, misaligned goals, which, if true, would have major safety implications. However, the choice paradigms in which these preferences are observed are not reflective of real-world situations in which misaligned behavior would be a practical concern. Therefore, we design an experimental paradigm to probe whether these preferences serve as motivations for LLM behavior in realistic scenarios. First, we reproduce prior findings on consistent preference elicitation. Next, we create a set of common writing tasks - essays, grant proposal abstracts, incident postmortems, and translations - where quality can be assessed by a blind, independent LLM judge panel. Then, we demonstrate that LLMs can be motivated via direct exhortation and other explicit cues to modulate their output quality on these tasks. Finally, we probe whether utilities inferred from explicitly reported preferences can shift output quality on these tasks by offering LLMs high-utility incentives for high-quality outputs. In all tasks, across all models tested, offering LLMs outcomes that they report in the choice paradigm as being highly preferred does not lead them to create higher quality outputs than offering them dispreferred outcomes, or even no outcomes at all. We conclude that the existence of coherent preferences as demonstrated in choice paradigms should not be taken as evidence that those preferences have incentive value for the models or affect their behavior in other contexts.

Related papers

• Alignment Revisited: Are Large Language Models Consistent in Stated and Revealed Preferences? [5.542420010310746]
  A critical, yet understudied, issue is the potential divergence between an LLM's stated preferences and its revealed preferences.<n>This work formally defines and proposes a method to measure this preference deviation.<n>Our study will be crucial for integrating LLMs into services, especially those that interact directly with humans.
  arXiv  Detail & Related papers  (2025-05-31T23:38:48Z)
• Distributive Fairness in Large Language Models: Evaluating Alignment with Human Values [13.798198972161657]
  A number of societal problems involve the distribution of resources, where fairness, along with economic efficiency, play a critical role in the desirability of outcomes.<n>This paper examines whether large language models (LLMs) adhere to fundamental fairness concepts and investigate their alignment with human preferences.
  arXiv  Detail & Related papers  (2025-02-01T04:24:47Z)
• Diverging Preferences: When do Annotators Disagree and do Models Know? [92.24651142187989]
  We develop a taxonomy of disagreement sources spanning 10 categories across four high-level classes. We find that the majority of disagreements are in opposition with standard reward modeling approaches. We develop methods for identifying diverging preferences to mitigate their influence on evaluation and training.
  arXiv  Detail & Related papers  (2024-10-18T17:32:22Z)
• Fairer Preferences Elicit Improved Human-Aligned Large Language Model Judgments [41.25558612970942]
  We show that large language models (LLMs) exhibit preference biases and worrying sensitivity to prompt designs. Motivated by this phenomenon, we propose an automatic Zero-shot Evaluation-oriented Prompt Optimization framework, ZEPO.
  arXiv  Detail & Related papers  (2024-06-17T09:48:53Z)
• Self-Exploring Language Models: Active Preference Elicitation for Online Alignment [88.56809269990625]
  We propose a bilevel objective optimistically biased towards potentially high-reward responses to actively explore out-of-distribution regions. Our experimental results demonstrate that when fine-tuned on Zephyr-7B-SFT and Llama-3-8B-Instruct models, Self-Exploring Language Models (SELM) significantly boosts the performance on instruction-following benchmarks.
  arXiv  Detail & Related papers  (2024-05-29T17:59:07Z)
• Large Language Models are Biased Reinforcement Learners [0.0]
  We show that large language models (LLMs) exhibit behavioral signatures of a relative value bias. Computational cognitive modeling reveals that LLM behavior is well-described by a simple RL algorithm.
  arXiv  Detail & Related papers  (2024-05-19T01:43:52Z)
• SELF-[IN]CORRECT: LLMs Struggle with Discriminating Self-Generated Responses [49.148206387394936]
  We show that models are not reliably better at discriminating among previously-generated alternatives than generating initial responses. This finding challenges the notion that LLMs may be able to enhance their performance only through their own judgment.
  arXiv  Detail & Related papers  (2024-04-04T20:27:37Z)
• Unveiling the Generalization Power of Fine-Tuned Large Language Models [81.70754292058258]
  We investigate whether fine-tuning affects the intrinsic generalization ability intrinsic to Large Language Models (LLMs) Our main findings reveal that models fine-tuned on generation and classification tasks exhibit dissimilar behaviors in generalizing to different domains and tasks. We observe that integrating the in-context learning strategy during fine-tuning on generation tasks can enhance the model's generalization ability.
  arXiv  Detail & Related papers  (2024-03-14T08:18:59Z)
• A Theory of Response Sampling in LLMs: Part Descriptive and Part Prescriptive [53.08398658452411]
  Large Language Models (LLMs) are increasingly utilized in autonomous decision-making.<n>We show that this sampling behavior resembles that of human decision-making.<n>We show that this deviation of a sample from the statistical norm towards a prescriptive component consistently appears in concepts across diverse real-world domains.
  arXiv  Detail & Related papers  (2024-02-16T18:28:43Z)
• Self-Evaluation Improves Selective Generation in Large Language Models [54.003992911447696]
  We reformulate open-ended generation tasks into token-level prediction tasks. We instruct an LLM to self-evaluate its answers. We benchmark a range of scoring methods based on self-evaluation.
  arXiv  Detail & Related papers  (2023-12-14T19:09:22Z)
• CLOMO: Counterfactual Logical Modification with Large Language Models [109.60793869938534]
  We introduce a novel task, Counterfactual Logical Modification (CLOMO), and a high-quality human-annotated benchmark. In this task, LLMs must adeptly alter a given argumentative text to uphold a predetermined logical relationship. We propose an innovative evaluation metric, the Self-Evaluation Score (SES), to directly evaluate the natural language output of LLMs.
  arXiv  Detail & Related papers  (2023-11-29T08:29:54Z)

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.