[2605.04180] MedFabric and EtHER: A Data-Centric Framework for Word-Level Fabrication Generation and Detection in Medical LLMs
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:2605.04180
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 > Computation and Language
arXiv:2605.04180 (cs)
[Submitted on 5 May 2026]
Title:MedFabric and EtHER: A Data-Centric Framework for Word-Level Fabrication Generation and Detection in Medical LLMs
Authors:Tung Sum Thomas Kwok, Qian Qian, Xiaofeng Lin, Dongxu Zhang, Jun Han, Zhichao Yang, Davin Hill, Tamer Soliman, Sanjit Singh Batra, Robert Tillman, Guang Cheng
View a PDF of the paper titled MedFabric and EtHER: A Data-Centric Framework for Word-Level Fabrication Generation and Detection in Medical LLMs, by Tung Sum Thomas Kwok and Qian Qian and Xiaofeng Lin and Dongxu Zhang and Jun Han and Zhichao Yang and Davin Hill and Tamer Soliman and Sanjit Singh Batra and Robert Tillman and Guang Cheng
View PDF
Abstract:Large Language Models exhibit strong reasoning and semantic understanding capabilities but often hallucinate in domains that require expert knowledge, among which fabrications, the generation of factually incorrect yet fluent statements, pose the greatest risk in medical contexts. Existing medical hallucination datasets inadequately capture fabrication phenomena due to limited fabrication coverage, stylistic disparities between human and LLM-authored texts, and distributional drift during hallucinated sample synthesis. To address this, we propose a data-centric pipeline to generate realistic and word-level fabrications that preserve syntactic and stylistic fidelity while introducing subtle factual deviations, resulting in MedFabric. Building upon this dataset, we introduce ETHER, a modular word-level fabrication detector integrating Text2Table Decomposition, Word Masking and Filling and Hybrid Sentence Pair Evaluation to enhance factual alignment. Empirical results demonstrate that MedFabric outperforms state-of-the-art detectors by over 15% on word-level fabrication benchmarks while maintaining consistent performance across structural similarities, offering a comprehensive framework for reliable and domain-specific factuality detection.
Subjects:
Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
Cite as: arXiv:2605.04180 [cs.CL]
(or arXiv:2605.04180v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2605.04180
Focus to learn more
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Tung Sum Thomas Kwok [view email] [v1] Tue, 5 May 2026 18:19:51 UTC (492 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled MedFabric and EtHER: A Data-Centric Framework for Word-Level Fabrication Generation and Detection in Medical LLMs, by Tung Sum Thomas Kwok and Qian Qian and Xiaofeng Lin and Dongxu Zhang and Jun Han and Zhichao Yang and Davin Hill and Tamer Soliman and Sanjit Singh Batra and Robert Tillman and Guang Cheng
View PDF
TeX Source
view license
Current browse context:
cs.CL
new | recent | 2026-05
Change to browse by:
cs cs.AI
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