{"ID":2871403,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.11355","arxiv_id":"2509.11355","title":"Promoting Shape Bias in CNNs: Frequency-Based and Contrastive Regularization for Corruption Robustness","abstract":"Convolutional Neural Networks (CNNs) excel at image classification but remain vulnerable to common corruptions that humans handle with ease. A key reason for this fragility is their reliance on local texture cues rather than global object shapes -- a stark contrast to human perception. To address this, we propose two complementary regularization strategies designed to encourage shape-biased representations and enhance robustness. The first introduces an auxiliary loss that enforces feature consistency between original and low-frequency filtered inputs, discouraging dependence on high-frequency textures. The second incorporates supervised contrastive learning to structure the feature space around class-consistent, shape-relevant representations. Evaluated on the CIFAR-10-C benchmark, both methods improve corruption robustness without degrading clean accuracy. Our results suggest that loss-level regularization can effectively steer CNNs toward more shape-aware, resilient representations.","short_abstract":"Convolutional Neural Networks (CNNs) excel at image classification but remain vulnerable to common corruptions that humans handle with ease. A key reason for this fragility is their reliance on local texture cues rather than global object shapes -- a stark contrast to human perception. To address this, we propose two c...","url_abs":"https://arxiv.org/abs/2509.11355","url_pdf":"https://arxiv.org/pdf/2509.11355v1","authors":"[\"Robin Narsingh Ranabhat\",\"Longwei Wang\",\"Amit Kumar Patel\",\"KC santosh\"]","published":"2025-09-14T17:14:07Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.AI\"]","methods":"[\"Convolutional Neural Network\"]","has_code":false}