Skip to content
AI News HubLIVE
Original source2 min read

Reference-Guided Data Curation for 3D Medical Vision-Language Pretraining

Summary

This paper proposes an offline, training-free, reference-guided data curation recipe that uses a small high-quality reference set to extract quality signals from a large uncurated pool of medical image-report pairs and merges them via rank aggregation for sample selection. Validated on 35k 3D abdominal CT and 58k 2D chest X-ray pairs, it matches or exceeds state-of-the-art zero-shot classification with 3.3× and up to 10× less training data, respectively.

SourcearXiv Computer VisionAuthor: Faidra Patsatzi, Ashwin Kumar, Chong Wang, Curtis P. Langlotz, Reinhard Heckel, Akshay S. Chaudhari
Reference-Guided Data Curation for 3D Medical Vision-Language Pretraining
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions
Read article

[Submitted on 17 Nov 2026]

Title:Reference-Guided Data Curation for 3D Medical Vision-Language Pretraining

View a PDF of the paper titled Reference-Guided Data Curation for 3D Medical Vision-Language Pretraining, by Faidra Patsatzi and 5 other authors

View PDF HTML (experimental)

Abstract:Training 3D vision-language models (VLMs) on large medical imaging datasets collected across institutions is constrained by data quality. Vast, uncurated pools of image-report pairs are characterized by redundancy, class imbalance, and cross-institutional heterogeneity. We address this challenge of data curation for improving 3D VLM pretraining by proposing an offline, training-free, reference-guided data curation recipe. Our approach leverages a small, high-quality reference set to independently extract quality signals for each modality of a large, uncurated pool and merges them through a rank-aggregation step into a single ranking that guides sample selection. We validate this method on datasets comprising 35k 3D abdominal CT volume-report pairs from two institutions and 58k 2D chest X-ray image-report pairs. On 3D CT, using our reference-guided data curation strategy, our curated model matches the zero-shot disease classification performance of a state-of-the-art baseline (Merlin) while using $3.3\times$ less training data and a reference set of 5k samples. On 2D chest X-rays, our method yields a curated model that matches or exceeds the zero-shot disease classification performance of a model trained on the full pool while using up to $10\times$ less training data and a reference set of only 74 samples. Ablations show that the selection is stable under resampling of the reference, that smaller reference sets still outperform random sampling, and that label rebalancing does not improve performance. To support reproducibility, we release our code.

Comments: Accepted at WACV 2027. Code: this https URL

Subjects:

Computer Vision and Pattern Recognition (cs.CV)

Cite as: arXiv:2611.23484 [cs.CV]

(or arXiv:2611.23484v1 [cs.CV] for this version)

https://doi.org/10.48550/arXiv.2611.23484

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Faidra Anastasia Patsatzi [view email] [v1] Tue, 17 Nov 2026 20:35:32 UTC (10,657 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled Reference-Guided Data Curation for 3D Medical Vision-Language Pretraining, by Faidra Patsatzi and 5 other authors

View PDF

HTML (experimental)

TeX Source

view license

Additional Features

Audio Summary

Current browse context:

cs.CV

new | recent | 2026-11

Change to browse by:

cs

References & Citations

SciX

Google Scholar

Semantic Scholar

Loading...

Data provided by:

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?)

Key points and analysis

Article intelligence

EngineersAdvanced

Key points

  • Proposes an offline, training-free reference-guided curation recipe that extracts per-modality quality signals from a small reference set and rank-aggregates them for sample selection
  • On 3D CT, matches Merlin's zero-shot disease classification with 3.3× less training data and a 5k-sample reference set
  • On 2D chest X-rays, matches or exceeds full-pool training with up to 10× less data and only 74 reference samples
  • Ablations show stable selection under reference resampling, smaller reference sets still beat random sampling, and label rebalancing does not help; code is released

Highlights and analysis are generated automatically and may contain errors. Check the original source.

Reference-Guided Data Curation for 3D Medical Vision-Language Pretraining | AI News Hub