Metraipolis › archive › dc_6ba91c9cac34
Mitigating Shortcut Learning: Texture-Penalized Prototype Networks — arXiv:2609.07504
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.07504
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T15:11:51.077Z
- SHA-256 of the archived text
58bc6703b8c8515de6d03ffc2085908cd5c85cae0893e86c49aa54d5101fdaf5- Stable URL of this page
https://metraipolis.com/documento/dc_6ba91c9cac34
How to cite
Mitigating Shortcut Learning: Texture-Penalized Prototype Networks — arXiv:2609.07504 — arXiv (en), https://arxiv.org/abs/2609.07504 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6ba91c9cac34 (Metraipolis archive dc_6ba91c9cac34, sha256 58bc6703b8c8515de6d03ffc2085908cd5c85cae0893e86c49aa54d5101fdaf5)
BibTeX
@misc{dc6ba91c9cac34,
title = {Mitigating Shortcut Learning: Texture-Penalized Prototype Networks — arXiv:2609.07504},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.07504},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6ba91c9cac34; sha256 58bc6703b8c8515de6d03ffc2085908cd5c85cae0893e86c49aa54d5101fdaf5},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.07504. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Standard Convolutional Neural Networks (CNNs) exhibit severe performance degradation due to a strong inductive texture bias that prioritizes local, high-frequency patterns over global structural shapes. This dependency causes confident misclassifications during textural changes or environmental effects. To address this flaw, this study introduces the Texture-Penalized Prototype Network (TPPN), a novel architectural framework that shifts this inherent bias without depending on resource-intensive augmented datasets. Specifically, a Texture-Penalization Branch (TPB) imposes a penalty to suppress the extraction of local texture proxies, forcing the network backbone to discard high-frequency cues and extract purified, shape-biased representations. By evaluating similarities within a prototype-based hypersphere derived from the final convolutional features, the approach enforces strict geometric constraints, treating objects as compositions of essential parts to achieve robust classification. Evaluations on texture-shape cue-conflict datasets and synthetic noise benchmarks demonstrate the stronger shape bias of this structural disentanglement. The proposed framework reduces the inherent texture bias of a baseline ResNet-50 from 55.11% to 29.73%, surpassing the texture-suppression capabilities of an off-the-shelf Vision Transformer (ViT-B/16). Furthermore, the approach demonstrates robust generalization under cue-conflict conditions, resisting textural shortcut learning when encountering Out-of-Distribution (OOD) shapes. The model maintains stronger shape accuracy against elevated perturbations. On clean validation data, the architecture incurs a minimal drop in accuracy of 0.90 percentage points. This provides a structural, efficient solution to CNN texture bias. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.07504 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-16T14:49:57.225Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).