@@ -0,0 +1,13 @@
+@article{dornaika2025deep,
+ abstract = {In machine learning, deep metric learning from original data is essential, with supervised contrastive learning being a notable approach. This method aims to form a deep feature space where similar samples from the same class are clustered together, while dissimilar samples from different classes are separated. However, a common limitation of contrastive learning methods is that they utilize the entire feature space for data embedding and often neglect the within-class variability. To overcome this limitation, we propose a novel supervised contrastive learning method that decomposes deep features into two distinct components: common features, which encapsulate the essential, class-defining characteristics, and style features, which capture the within-class variability and nuanced differences. Additionally, we enhance this approach by introducing an overlapping field that synergistically integrates elements from both feature spaces, enabling a more comprehensive and robust feature representation. Our experiments with different image datasets and deep encoders, including CNNs and transformers, show that our approach outperforms traditional single-feature contrastive methods. On the CIFAR100 and PASCAL VOC databases, traditional supervised contrastive learning achieved accuracy rates of 75.5% and 51.41%, respectively, while our method improved them to 77.81% and 59.38%, respectively. We present an algorithm for deep contrastive learning that utilizes two feature spaces: one for encoding common class features and another for capturing within-class variability. This is achieved by partitioning the features of the last layer of the encoder into (i) a common field and (ii) a style field. Our loss function contrasts the common features while summarizing the style features within the same class so that the style field can capture the intra-class variability.},
+ author = {Fadi Dornaika and Bin Wang and Jinan Charafeddine},
+ doi = {10.1007/s12559-025-10430-4},
+ journal = {Cognitive Computation},
+ number = {3},
+ pages = {117},
+ publisher = {Springer},
+ title = {Deep feature disentanglement for supervised contrastive learning: Application to image classification},
+ url = {https://link.springer.com/article/10.1007/s12559-025-10430-4},
+ volume = {17},
+ year = {2025}
+}
@@ -0,0 +1,40 @@
+---
+title: 'Deep feature disentanglement for supervised contrastive learning: Application
+  to image classification'
+authors:
+- Fadi Dornaika
+- Bin Wang
+- Jinan Charafeddine
+date: '2025-01-01'
+publishDate: '2026-04-14T13:10:00.966307Z'
+publication_types:
+- article-journal
+publication: '*Cognitive Computation*'
+doi: 10.1007/s12559-025-10430-4
+abstract: 'In machine learning, deep metric learning from original data is essential,
+  with supervised contrastive learning being a notable approach. This method aims
+  to form a deep feature space where similar samples from the same class are clustered
+  together, while dissimilar samples from different classes are separated. However,
+  a common limitation of contrastive learning methods is that they utilize the entire
+  feature space for data embedding and often neglect the within-class variability.
+  To overcome this limitation, we propose a novel supervised contrastive learning
+  method that decomposes deep features into two distinct components: common features,
+  which encapsulate the essential, class-defining characteristics, and style features,
+  which capture the within-class variability and nuanced differences. Additionally,
+  we enhance this approach by introducing an overlapping field that synergistically
+  integrates elements from both feature spaces, enabling a more comprehensive and
+  robust feature representation. Our experiments with different image datasets and
+  deep encoders, including CNNs and transformers, show that our approach outperforms
+  traditional single-feature contrastive methods. On the CIFAR100 and PASCAL VOC databases,
+  traditional supervised contrastive learning achieved accuracy rates of 75.5% and
+  51.41%, respectively, while our method improved them to 77.81% and 59.38%, respectively.
+  We present an algorithm for deep contrastive learning that utilizes two feature
+  spaces: one for encoding common class features and another for capturing within-class
+  variability. This is achieved by partitioning the features of the last layer of
+  the encoder into (i) a common field and (ii) a style field. Our loss function contrasts
+  the common features while summarizing the style features within the same class so
+  that the style field can capture the intra-class variability.'
+links:
+- name: URL
+  url: https://link.springer.com/article/10.1007/s12559-025-10430-4
+---
@@ -0,0 +1,13 @@
+@article{wang2026scs,
+ abstract = {Image classification can be inherently challenging due to subtle inter-class differences and substantial intra-class variations, limiting the effectiveness of existing contrastive learning approaches. In particular, supervised contrastive methods based on the InfoNCE loss often suffer from the negative-sample dilution issue and lack explicit mechanisms for adaptive decision-boundary control, significantly weakening their discriminative capability on fine-grained image classification tasks. To address these challenges, we propose a novel supervised contrastive learning framework, termed Sigmoid-based Common and Style Supervised Contrastive Learning (SCS-SupCon). In this framework, we introduce a sigmoid-based pairwise contrastive loss with adaptive decision boundaries, explicitly parameterized by learnable temperature and bias terms. This design places greater emphasis on critical discriminative information from hard negatives, thereby alleviating the problem of negative-sample dilution while fully leveraging supervision signals in contrastive learning. Furthermore, we incorporate an explicit style-distance constraint to disentangle style and content representations, leading to more robust and discriminative feature learning. Comprehensive experiments on six benchmark datasets, including prominent fine-grained datasets such as CUB200-2011 and Stanford Dogs, consistently demonstrate that our proposed SCS-SupCon achieves superior performance over the most closely related InfoNCE-based supervised contrastive baselines (SupCon, SelfCon, CS-SupCon and its overlapping variant) across diverse CNN and Transformer backbones. In particular, on CIFAR-100 with a ResNet-50 encoder, SCS-SupCon improves the mean top-1 accuracy over SupCon by about 3.9 percentage points and over CS-SupCon by about 1.7 percentage points under a five-fold cross-validation protocol. On challenging fine-grained datasets such as CUB200-2011 and Stanford Dogs with both CNN and Transformer architectures, our method achieves absolute improvements of approximately 0.4–3.0 percentage points over CS-SupCon. Extensive ablation studies and paired statistical tests further confirm the robustness and effectiveness of our framework, and a Friedman test with Nemenyi post-hoc analysis shows that SCS-SupCon attains the best average rank among the evaluated methods, even though pairwise differences with other strong competitors are not always statistically significant at the 0.05 level.},
+ author = {Bin Wang and Fadi Dornaika and Jinan Charafeddine},
+ doi = {https://doi.org/10.1016/j.eswa.2026.131308},
+ issn = {0957-4174},
+ journal = {Expert Systems with Applications},
+ keywords = {Supervised contrastive learning, Fine-grained classification, Negative sample dilution, Feature disentanglement, Decision boundary adjustment},
+ pages = {131308},
+ title = {SCS-SupCon: Sigmoid-based common and style supervised contrastive learning with adaptive decision boundaries},
+ url = {https://www.sciencedirect.com/science/article/pii/S0957417426002216},
+ volume = {311},
+ year = {2026}
+}
@@ -0,0 +1,53 @@
+---
+title: 'SCS-SupCon: Sigmoid-based common and style supervised contrastive learning
+  with adaptive decision boundaries'
+authors:
+- Bin Wang
+- Fadi Dornaika
+- Jinan Charafeddine
+date: '2026-01-01'
+publishDate: '2026-04-14T13:10:00.976472Z'
+publication_types:
+- article-journal
+publication: '*Expert Systems with Applications*'
+doi: https://doi.org/10.1016/j.eswa.2026.131308
+abstract: Image classification can be inherently challenging due to subtle inter-class
+  differences and substantial intra-class variations, limiting the effectiveness of
+  existing contrastive learning approaches. In particular, supervised contrastive
+  methods based on the InfoNCE loss often suffer from the negative-sample dilution
+  issue and lack explicit mechanisms for adaptive decision-boundary control, significantly
+  weakening their discriminative capability on fine-grained image classification tasks.
+  To address these challenges, we propose a novel supervised contrastive learning
+  framework, termed Sigmoid-based Common and Style Supervised Contrastive Learning
+  (SCS-SupCon). In this framework, we introduce a sigmoid-based pairwise contrastive
+  loss with adaptive decision boundaries, explicitly parameterized by learnable temperature
+  and bias terms. This design places greater emphasis on critical discriminative information
+  from hard negatives, thereby alleviating the problem of negative-sample dilution
+  while fully leveraging supervision signals in contrastive learning. Furthermore,
+  we incorporate an explicit style-distance constraint to disentangle style and content
+  representations, leading to more robust and discriminative feature learning. Comprehensive
+  experiments on six benchmark datasets, including prominent fine-grained datasets
+  such as CUB200-2011 and Stanford Dogs, consistently demonstrate that our proposed
+  SCS-SupCon achieves superior performance over the most closely related InfoNCE-based
+  supervised contrastive baselines (SupCon, SelfCon, CS-SupCon and its overlapping
+  variant) across diverse CNN and Transformer backbones. In particular, on CIFAR-100
+  with a ResNet-50 encoder, SCS-SupCon improves the mean top-1 accuracy over SupCon
+  by about 3.9 percentage points and over CS-SupCon by about 1.7 percentage points
+  under a five-fold cross-validation protocol. On challenging fine-grained datasets
+  such as CUB200-2011 and Stanford Dogs with both CNN and Transformer architectures,
+  our method achieves absolute improvements of approximately 0.4–3.0 percentage points
+  over CS-SupCon. Extensive ablation studies and paired statistical tests further
+  confirm the robustness and effectiveness of our framework, and a Friedman test with
+  Nemenyi post-hoc analysis shows that SCS-SupCon attains the best average rank among
+  the evaluated methods, even though pairwise differences with other strong competitors
+  are not always statistically significant at the 0.05 level.
+tags:
+- Supervised contrastive learning
+- Fine-grained classification
+- Negative sample dilution
+- Feature disentanglement
+- Decision boundary adjustment
+links:
+- name: URL
+  url: https://www.sciencedirect.com/science/article/pii/S0957417426002216
+---