[2605.05329] Understanding Annotator Safety Policy with Interpretability
Skip to main content
Learn about arXiv becoming an independent nonprofit.
We gratefully acknowledge support from the Simons Foundation, member institutions, and all contributors. Donate
> cs > arXiv:2605.05329
Help | Advanced Search
All fields Title Author Abstract Comments Journal reference ACM classification MSC classification Report number arXiv identifier DOI ORCID arXiv author ID Help pages Full text
Search
GO
quick links
Login
Help Pages
About
-->
Computer Science > Artificial Intelligence
arXiv:2605.05329 (cs)
[Submitted on 6 May 2026]
Title:Understanding Annotator Safety Policy with Interpretability
Authors:Alex Oesterling, Donghao Ren, Yannick Assogba, Dominik Moritz, Sunnie S.Y. Kim, Leon Gatys, Fred Hohman
View a PDF of the paper titled Understanding Annotator Safety Policy with Interpretability, by Alex Oesterling and 6 other authors
View PDF HTML (experimental)
Abstract:Safety policies define what constitutes safe and unsafe AI outputs, guiding data annotation and model development. However, annotation disagreement is pervasive and can stem from multiple sources such as operational failures (annotators misunderstand or misexecute the task), policy ambiguity (policy wording leaves room for interpretation), or value pluralism (different annotators hold different perspectives on safety). Distinguishing these sources matters. For example, operational failures call for quality control, ambiguity calls for policy clarification, and pluralism calls for deliberation about incorporating diverse perspectives. Yet understanding why annotators disagree is difficult. Directly asking annotators for their reasoning is costly, substantially increasing annotation burden, and can be unreliable for both human and LLM annotators as self-reported reasoning often fails to reflect actual decision processes.
We introduce Annotator Policy Models (APMs), interpretable models that learn annotators' internal safety policies from labeling behavior alone, making annotator reasoning visible and comparable without additional annotation effort. We validate that APMs accurately model annotator safety policy (>80% accuracy), faithfully predict responses to counterfactual edits, and recover known policy differences in controlled settings. Applying APMs to LLM and human annotations, we demonstrate two core applications: (1) surfacing policy ambiguity by revealing how annotators interpret safety instructions differently, and (2) surfacing value pluralism by uncovering systematic differences in safety priorities across demographic groups. Together, these capabilities support more targeted, transparent, and inclusive safety policy design.
Comments: 38 pages, 13 figures, ACM FAccT 2026
Subjects:
Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Cite as: arXiv:2605.05329 [cs.AI]
(or arXiv:2605.05329v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2605.05329
Focus to learn more
arXiv-issued DOI via DataCite
Related DOI:
https://doi.org/10.1145/3805689.3806472
Focus to learn more
DOI(s) linking to related resources
Submission history
From: Alex Oesterling [view email] [v1] Wed, 6 May 2026 18:01:40 UTC (6,628 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled Understanding Annotator Safety Policy with Interpretability, by Alex Oesterling and 6 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.AI
new | recent | 2026-05
Change to browse by:
cs cs.LG
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?)
About
Help
contact arXivClick here to contact arXiv Contact
subscribe to arXiv mailingsClick here to subscribe Subscribe
Copyright
Privacy Policy
Web Accessibility Assistance
arXiv Operational Status