资讯

From Preference Data to Personalization: Tracing Sycophancy in Large Language Models

📅 2026-11-18 ⏱️ 约 6 分钟阅读 ✍️ AI导航编辑部 🔗 arxiv.org
From Preference Data to Personalization: Tracing Sycophancy in Large Language Models
📝 内容摘要

Computer Science > Computation and Language [Submitted on 15 Oct 2026] Title:From Preference Data to Personalizati

📌 核心要点

  • Computer Science > Computation and Language
  • [Submitted on 15 Oct 2026]
  • From: Himaghna Bhattacharjee [view email][v1] Thu, 15 Oct 2026 04:53:04 UTC (463 KB)

Computer Science > Computation and Language

[Submitted on 15 Oct 2026]

Title:From Preference Data to Personalization: Tracing Sycophancy in Large Language Models

View PDF HTML (experimental)Abstract:Sycophancy in large language models is often measured as agreement with the user, yet agreement can be appropriate when the user's claim is valid, their preference is genuine, or their experience warrants acknowledgment. We study sycophancy as inappropriate agreement: validating a claim, preference, or self-assessment when the context calls for correction, qualification, or resistance. We introduce a validity-aware sycophancy grader and SycoDrift, a persona-targeted evaluation set of 3,298 validity-annotated probes with controlled annotations for user openness and stakes. We use this framework to analyze three sources of agreement pressure: human preference data, automated judge specification, and in-context personalization. First, preference for validating responses in human datasets (Chatbot Arena, HH-RLHF, Community Alignment) is weak and concentrated where agreement is often appropriate. Second, in a selected-value, single-model experiment, emphasized judge rubrics produce consistent but modest increases relative to paired suppressed rubrics, demonstrating a plausible mechanism for training-time amplification. Third, in off-policy evaluations across 14 models, inference-time personalization increases inappropriate agreement, with the largest effect from matched user profiles placed directly before the query.

Submission history

From: Himaghna Bhattacharjee [view email][v1] Thu, 15 Oct 2026 04:53:04 UTC (463 KB)

Additional Features

References & Citations

Loading...

Bibliographic and Citation Tools

Bibliographic Explorer (What is the Explorer?)

Connected Papers (What is Connected Papers?)

Litmaps (What is Litmaps?)

scite Smart Citations (What are Smart Citations?)

Code, Data and Media Associated with this Article

alphaXiv (What is alphaXiv?)

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub (What is DagsHub?)

Gotit.pub (What is GotitPub?)

Hugging Face (What is Huggingface?)

ScienceCast (What is ScienceCast?)

Demos

Recommenders and Search Tools

Influence Flower (What are Influence Flowers?)

CORE Recommender (What is CORE?)

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.

来源arxiv.org· 本文为编辑整理,仅供参考