Papers/2608.18185
🧪 Test?View on arXiv

H$^2$EDL: Hyper Evidential Deep Learning for Hierarchical Classification

Not specified in the provided content

hierarchical classificationevidential reasoninguncertainty quantification
2608.18185
Builder Relevance
70%
1h ago

Abstract

H$^2$EDL introduces a model that effectively captures uncertainty in hierarchical classification by combining evidential and hierarchical approaches.

Reality Card

Core Claim

H$^2$EDL reduces calibration error by approximately half compared to cross-entropy baselines, especially at deeper hierarchy levels.

Method / Result

Calibration error reduction by approximately 50% on FGVC-Aircraft and DERM12345 datasets.

Limitations

The model's reliance on the taxonomy structure may limit its applicability to datasets without a clear hierarchical organization.

Paper to code

Verified implementation resources so builders can test the paper’s claims instead of stopping at the abstract.

No verified implementation link has been attached yet. AIBuzzHub will keep this panel separate from unverified search results.
← Back to all papers