论文研究
arxiv.org ·
为人格感知大语言模型交互创建原子用户模型
该 arXiv 预印本提出 Atomic User Model（AUM）：一个以稳定身份核心为中心、覆盖心理、认知、经验、行为与社会层面的可读用户画像，并描述所谓“人格渗漏”现象——提示词措辞会携带写作者的人格痕迹并被助手无意识模仿。作者不将完整画像插入提示，而是把 AUM 作为可检索索引，通过任务分类器、字段选择与预算受限的检索器仅传入少量相关字段。在 16 个模拟用户、6 个风格敏感任务、3 个随机种子上测试：8 个检索字段即可达到完整画像的写作质量，仅用 23% 的上下文（211 token 对 915 token），比普通偏好笔记高 0.24 分（5 分制）；从 4 个样本中识别用户本人写作的准确率从 14.9% 升至 42.7%（随机猜测为 25%）。四项预注册对照实验无显著效应，说明增益来自画像结构而非检索方法。
AI推荐理由 arXiv 预印本，提出 Atomic User Model（AUM）用于人格感知的 LLM 交互，并给出可验证的量化结果：仅用 8 个检索字段（211 token，占完整画像 915 token 的 23%）即可达到完整画像的写作质量，比普通偏好笔记高 0.24 分（5 分制），从 4 个样本中识别用户本人写作的准确率由 14.9% 升至 42.7%（随机猜测为 25%）。局限在于实验基于 16 个模拟用户、6 个任务、3 个随机种子，尚未在真实用户上验证，且论文为预印本未经同行评审。
72
正文
Computer Science > Human-Computer Interaction
arXiv:2609.12086 (cs)
[Submitted on 10 Sep 2026 (v1), last revised 16 Sep 2026 (this version, v2)]
Title:Creating an Atomic User Model for Personality-Aware Large Language Model Interaction
Authors:B. Sankar, Deepthika S, Pawni Yadav, Amogh A S
View a PDF of the paper titled Creating an Atomic User Model for Personality-Aware Large Language Model Interaction, by B. Sankar and 3 other authors
View PDF
HTML (experimental)
Abstract:Assistants built on large language models are expected to write in their users' own voice. Most systems summarise the user's preferences and include the summary in the prompt. This is the wrong way round. Preferences are only the surface of a person and change with the task, while the underlying personality stays the same, so storing preferences alone means relearning the user afresh whenever the task changes. This paper makes four contributions. First, we describe an effect we call personality seepage: the wording of a prompt carries traces of the writer's personality, which the assistant copies without knowing the writer. Second, we propose the Atomic User Model (AUM), a readable profile with a stable identity core surrounded by four layers covering psychological, cognitive, experiential, behavioral, and social details, plus notes on inner conflict and authenticity. Third, instead of inserting the entire profile, we use AUM as a searchable index, in which a task classifier, a selection step, and a budgeted retriever pass along only a few relevant fields. Fourth, we test the pipeline with 16 simulated users, 6 style-sensitive tasks, and 3 seeds. Eight retrieved fields matched the writing quality of the whole profile, while using only 23 percent of the context (211 tokens instead of 915). They scored 0.24 points higher than a plain preference note on a five-point scale. Accuracy in picking a user's own writing from four samples rose from 14.9 to 42.7 percent, where guessing gives 25 percent. Four pre-registered controls showed no effect, so the gain comes from the profile's structure rather than the search method. Personalization helps most for the users for whom a generic assistant imitates them the worst.
Comments:
Subjects:
Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
MSC classes:
68T50, 68T01, 68T05
ACM classes:
I.2.7; H.5.2; I.2.11
Cite as:
arXiv:2609.12086 [cs.HC]
(or
arXiv:2609.12086v2 [cs.HC] for this version)
https://doi.org/10.48550/arXiv.2609.12086
Focus to learn more
arXiv-issued DOI via DataCite
Submission history
From: Sankar B [view email]
[v1]
Thu, 10 Sep 2026 18:11:43 UTC (18,363 KB)
[v2]
Wed, 16 Sep 2026 10:15:54 UTC (18,363 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled Creating an Atomic User Model for Personality-Aware Large Language Model Interaction, by B. Sankar and 3 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.HC
new
|
recent
| 2026-09
Change to browse by:
cs
cs.AI
cs.CL
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?)