Fragile Preferences: A Deep Dive Into Order Effects in Large Language Models

[2506.14092] Fragile Preferences: A Deep Dive Into Order Effects in Large Language Models Skip to main content Search Submit Donate Log in Search arXiv Press Enter to search · Advanced search Computer Science > Artificial Intelligence arXiv:2506.14092 (cs) [Submitted on 17 Jun 2025 ( v1 ), last revised 23 Jul 2026 (this version, v4)] Title: Fragile Preferences: A Deep Dive Into Order Effects in Large Language Models Authors: Haonan Yin , Shai Vardi , Vidyanand Choudhary View a PDF of the paper titled Fragile Preferences: A Deep Dive Into Order Effects in Large Language Models, by Haonan Yin and 2 other authors View PDF HTML (experimental) Abstract: Large language models (LLMs) are increasingly deployed in decision-support systems for high-stakes domains such as hiring and university admissions, where choices often involve selecting among competing alternatives. While prior work has noted position biases in LLM-driven comparisons, these biases have not been systematically analyzed or linked to underlying preference structures. We present the first comprehensive study of position biases across multiple LLMs and two distinct domains: resume comparisons, representing a realistic high-stakes context, and color selection, which isolates position effects by removing confounding factors. We find strong and consistent order effects, including a quality-dependent shift: when all options are high quality, models favor the first option, but when quality is lower, they favor later options. We also identify a previously undocumented bias: a name bias, where certain names are favored despite controlling for demographic signals. To separate superficial tie-breaking from genuine distortions of judgment, we extend the rational choice framework to classify pairwise preferences as robust, fragile, or indifferent. Using this framework, we show that order effects can lead models to select strictly inferior options. These results indicate that LLMs exhibit distinct failure modes not documented in human decision-making. We also propose targeted mitigation strategies, including a novel use of the temperature parameter, to recover underlying preferences when order effects distort model behavior. Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2506.14092 [cs.AI] (or arXiv:2506.14092v4 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2506.14092 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Haonan Yin [ view email ] [v1] Tue, 17 Jun 2025 01:14:22 UTC (6,100 KB) [v2] Sun, 17 Aug 2025 03:47:06 UTC (5,248 KB) [v3] Tue, 14 Apr 2026 04:37:42 UTC (6,539 KB) [v4] Thu, 23 Jul 2026 04:06:28 UTC (6,560 KB) Full-text links: Access Paper: View a PDF of the paper titled Fragile Preferences: A Deep Dive Into Order Effects in Large Language Models, by Haonan Yin and 2 other authors View PDF HTML (experimental) TeX Source view license Current browse context: cs.AI < prev | next > new | recent | 2025-06 Change to browse by: cs References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer ( What is the Explorer? ) Connected Papers Toggle Connected Papers ( What is Connected Papers? ) Litmaps Toggle Litmaps ( What is Litmaps? ) scite.ai Toggle scite Smart Citations ( What are Smart Citations? ) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv ( What is alphaXiv? ) Links to Code Toggle CatalyzeX Code Finder for Papers ( What is CatalyzeX? ) DagsHub Toggle DagsHub ( What is DagsHub? ) GotitPub Toggle Gotit.pub ( What is GotitPub? ) Huggingface Toggle Hugging Face ( What is Huggingface? ) ScienceCast Toggle ScienceCast ( What is ScienceCast? ) Demos Demos Replicate Toggle Replicate ( What is Replicate? ) Spaces Toggle Hugging Face Spaces ( What is Spaces? ) Spaces Toggle TXYZ.AI ( What is TXYZ.AI? ) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower ( What are Influence Flowers? ) Core recommender toggle CORE Recommender ( What is CORE? ) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs . Which authors of this paper are endorsers? | Disable MathJax ( What is MathJax? ) We gratefully acknowledge support from our major funders , member institutions , , and all contributors. About · Help · Contact · Subscribe · Copyright · Privacy · Accessibility · Operational Status (opens in new tab) Major funding support from

Related documents

• Attention-Path Fragility as an Uncertainty Signal in Large Language Models#444662
• Neural Collapse Is Forbidden: Information Floors in Language Models#361445
• Hidden in Plain Sight: The Overlooked Significance of Canonical Elements for Extreme LLM Sparsity#974671
• Beyond Right and Wrong: Evaluating Second-order Social Reasoning in Large Language Models#276804
• MulDimIF: A Multi-Dimensional Constraint Framework for Evaluating and Improving Instruction Following in Large Language Models#144286