Skip to content
AI News HubLIVE
Original source2 min read

Gradient-Based Latent Decomposition Reveals Mechanisms of Feature Degradation in Weakly Supervised Mammography

Summary

Researchers introduce gradient-based orthogonal latent decomposition for hierarchical VAEs to explain why weakly supervised mammography models preserve coarse lesion-type features under reconstruction while fine-grained malignancy cues degrade. On 3,550 CBIS-DDSM ROIs, only about 4.4% of latent magnitude aligns with supervisory gradients, leaving roughly 95.6% in an orthogonal residual that fine-grained pathology prediction depends on.

SourcearXiv Computer VisionAuthor: Vinceline Bertrand, Ionut Cardei
Gradient-Based Latent Decomposition Reveals Mechanisms of Feature Degradation in Weakly Supervised Mammography
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 24 Jul 2026]

Title:Gradient-Based Latent Decomposition Reveals Mechanisms of Feature Degradation in Weakly Supervised Mammography

View a PDF of the paper titled Gradient-Based Latent Decomposition Reveals Mechanisms of Feature Degradation in Weakly Supervised Mammography, by Vinceline Bertrand and Ionut Cardei

View PDF HTML (experimental)

Abstract:Weakly supervised hierarchical models exhibit a persistent asymmetry: coarse lesion-type features are preserved under reconstruction while fine-grained malignancy cues degrade---a pattern with direct consequences for the clinical reliability of breast cancer screening pipelines. We introduce gradient-based orthogonal latent decomposition for hierarchical Variational Autoencoders~(H-VAEs) to mechanistically explain this asymmetry. The latent space is partitioned into a task-aligned component~($z_1$), shaped by coarse supervisory gradients, and an orthogonal residual~($z_{\text{res}}$) capturing remaining representational capacity. On~3,550 mammographic Regions of Interest~(ROIs) from CBIS-DDSM, only~$\sim$4.4\% of latent magnitude aligns with supervisory gradients, leaving~$\sim$95.6\% in the orthogonal residual upon which fine-grained pathology prediction primarily depends. The model achieves Stage-1~AUC~0.866 and Stage 2~AUC~0.552, with a reconstruction stability gap of $\Delta_{\text{diag}}=5\%$ ($p=0.005$) and a classification gap of $\Delta_{\text{AUC}}=0.314$ ($p{

new | recent | 2026-07

Change to browse by:

cs cs.LG

References & Citations

NASA ADS

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

ResearchersAdvanced

Key points

  • The method partitions the H-VAE latent space into a task-aligned component z1 shaped by coarse supervisory gradients and an orthogonal residual zres holding the remaining representational capacity.
  • On 3,550 mammographic ROIs from CBIS-DDSM, only ~4.4% of latent magnitude aligns with supervisory gradients, while ~95.6% sits in the orthogonal residual.
  • The model reaches Stage-1 AUC 0.866 and Stage-2 AUC 0.552, with a 5% reconstruction stability gap (p=0.005) and a 0.314 classification AUC gap (p<0.001).
  • Comparisons with multi-instance and multi-task learning suggest the findings generalize across architectures and modalities.

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