@article{allaouiLGTargetSegDeepSemantic2027,

title = {LG-TargetSeg: Deep semantic fusion via language-conditioned selective scanning for targeted medical image segmentation},

author = {M. L. Allaoui and M. S. Allili},

url = {https://www.scopus.com/pages/publications/105051242411?origin=resultslist},

doi = {10.1016/j.inffus.2026.104797},

issn = {15662535 (ISSN)},

year  = {2027},

date = {2027-01-01},

journal = {Information Fusion},

volume = {139},

publisher = {Elsevier B.V.},

abstract = {Clinical language defines the semantic intent of targeted medical image segmentation, yet most vision–language methods incorporate it only at a shallow level or through training-time conditioning, leaving spatial reasoning largely language-blind. We propose LG-TargetSeg (Language-Guided Targeted Segmentation), a cross-modal framework that embeds clinical language directly into the selective scan of a Mamba-based encoder-decoder. Natural-language prompts, encoded by a frozen MedSigLIP foundation model, shape selective scanning throughout the encoder–decoder and additionally adapt the bottleneck state transition, alongside gated modulation and cross-attention, enabling semantic guidance to continuously shape feature propagation toward the target anatomy. A Text-Driven Selective Scan 2D (TD-SS2D) module achieves prompt-aware multi-scale fusion with linear complexity in the number of spatial tokens. To improve robustness to boundary ambiguity and class imbalance, we introduce the Reinforced Gaussian Dice Loss (RGDL) with learnable weighting parameters. Evaluated under both task-specific and universal generalization settings, LG-TargetSeg achieves state-of-the-art performance. Moreover, its prompt-driven selectivity remains reliable under genuine multi-target competition: on abdominal CT images containing up to 16 annotated organs, changing only the text prompt on a frozen model produces organ masks with less than 1% pairwise overlap. Additional materials and code are available at: https://github.com/MOHAMEDLamine77/LG-TargetSeg . © 2026 Elsevier B.V.},

keywords = {Computerized tomography, decoding, Deep vision-language fusion, Encoder-decoder, Gaussians, Medical image processing, Medical image segmentation, Modeling languages, Natural language processing systems, Reinforced gaussian dice loss, Selective scanning, Semantic fusion, Semantic segmentation, Semantics, Shallow levels, Signal encoding, State space methods, State-space modeling, State-space models, Training time},

pubstate = {published},

tppubtype = {article}

}

@article{ledrou-paquetFacialInformationUnderlying2026a,

title = {The facial information underlying economic decision-making},

author = {V. Ledrou-Paquet and D. Fiset and M. Carré and J. Guérette and C. Blais},

url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-105020458633&doi=10.1177%2F03010066251387848&partnerID=40&md5=7c1ead96fa13944073ed63bc952ca723},

doi = {10.1177/03010066251387848},

issn = {03010066 (ISSN)},

year  = {2026},

date = {2026-01-01},

journal = {Perception},

volume = {55},

number = {3},

pages = {243–265},

abstract = {Faces are rapidly and automatically assessed on multiple social dimensions, including trustworthiness. The high inter-rater agreement on this social judgment suggests a systematic association between facial appearance and perceived trustworthiness. The facial information used by observers during explicit trustworthiness judgments has been studied before. However, it remains unknown whether the same perceptual strategies are used during decisions that involve trusting another individual, without necessitating an explicit trustworthiness judgment. To explore this, 53 participants completed the Trust Game, an economic decision task, while facial information was randomly sampled using the Bubbles method. Our results show that economic decisions based on facial cues rely on similar visual information as that used during explicit trustworthiness judgments. We then manipulated facial features identified as diagnostic for trust to test their influence on perceived trustworthiness (Experiment 2) and on trust-related behaviors (Experiment 3). Across all experiments, subtle, targeted changes to facial features systematically shifted both impressions and monetary trust decisions. These findings demonstrate that the same perceptual strategies underlie explicit judgments and trust behaviors, highlighting the applied relevance of even minimal alterations in facial appearance. These findings should be replicated with real faces from diverse demographic backgrounds to confirm their generalizability. © The Author(s) 2025. This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 License (https://creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access page (https://us.sagepub.com/en-us/nam/open-access-at-sage).},

keywords = {Bubbles, face evaluation, face perception, facial trustworthiness, Social Perception, trust game, trustworthiness},

pubstate = {published},

tppubtype = {article}

}

@article{monthuy-blancExaminingDistinctivenessBody2026a,

title = {Examining the distinctiveness of body image and self-related constructs in eating disorders using virtual reality: the role of shape concerns, physical self-worth, and global self-worth},

author = {J. Monthuy-Blanc and G. Fortin and G. Corno and S. Bouchard},

url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-105029796609&doi=10.3389%2Ffpsyg.2025.1649698&partnerID=40&md5=6051a2b03589f05fea165ff9d1d84200},

doi = {10.3389/fpsyg.2025.1649698},

issn = {16641078 (ISSN)},

year  = {2026},

date = {2026-01-01},

journal = {Frontiers in Psychology},

volume = {16},

abstract = {Introduction: In Western culture, the female body is commonly socially perceived as an object of evaluation, causing women to frequently evaluate their self-worth based on their physical appearance. Since the last decade, the use of virtual reality (VR) helped clarify the intricate interplay between broader self-related dimensions and dysfunctional eating attitudes and behaviors in women with eating disorders (EDs). The first objective of this study explores the role of body image attitudes (i.e., perceived physical attractiveness, body shape concerns), global self-worth, and physical self-worth in determining visual-perceptual body image representations (i.e., allocentric and egocentric ideal and self-perceived body size) and visual-perceptual body image disturbances (i.e., allocentric and egocentric visual-perceptual body dissatisfaction) in a sample of women with EDs. Additionally, the second objective is to explore the role of body image variables (in terms of attitudes: perceived physical attractiveness, body shape concerns; and in terms of visual-perceptual body image disturbances), global self-worth, physical self-worth, in determining dysfunctional eating attitudes and behaviors (i.e., eating concerns, restraint, and bulimia) in women with EDs. Methods: The sample involved 96 self-identified female participants. Pearson's bivariate correlations and multiple linear regressions analyses were conducted to investigate the study's objectives. A VR-based figure rating scale was used to perform visual depictive body size estimation tasks in an allocentric and egocentric perspectives. Results: The findings indicate that physical self-worth and shape concerns are the primary variables related to visual-perceptual body image dissatisfaction. Shape concerns is also associated with eating concerns. Discussion: This study emphasizes the critical role of physical self-worth and shape concerns as common variables of interest in relation to both visual-perceptual body image representations and disturbances, as well as shape concerns for dysfunctional eating attitudes and behaviors. These findings clarify the understating of the intricate interplay between body image, broader self-related dimensions, and dysfunctional eating attitudes and behaviors in EDs. Copyright © 2026 Monthuy-Blanc, Fortin, Corno and Bouchard.},

keywords = {body dissatisfaction, dysfunctional eating attitude and behaviors, eating concerns, eating pathology, physical self-perceptions, self-esteem, shape concerns, virtual reality},

pubstate = {published},

tppubtype = {article}

}

@article{alessandraPreselectedPreferredImmersive2026a,

title = {Preselected and preferred immersive virtual reality versus narrative alone to induce post-stress relaxation in patients with pulmonary arterial hypertension: A pilot study on perceived stress and heart rate},

author = {G. Alessandra and D. M. Beatrice and C. Sara and M. Roberta and V. Martina and K. Patrycja and B. Maurizio and B. Luca and D. V. Laura Adelaide and B. Stéphane},

url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-105029595132&doi=10.1111%2Fbjhp.70059&partnerID=40&md5=b86f0151066a7923d8854913cbd54318},

doi = {10.1111/bjhp.70059},

issn = {1359107X (ISSN)},

year  = {2026},

date = {2026-01-01},

journal = {British Journal of Health Psychology},

volume = {31},

number = {1},

abstract = {Objectives: Several studies have shown the effectiveness of immersive virtual reality (IVR)-based relaxation techniques in alleviating stress within the general population. However, few data are available on patients, or on the effectiveness of different scenarios in inducing relaxation. This pilot study aims to compare the effectiveness of three relaxation techniques—preselected IVR (IVR-PS), preferred IVR (i.e. chosen by the participant from different alternatives—IVR-PR), and narrative alone (CTR)—in reducing physiological and psychological stress in 16 pulmonary arterial hypertension (PAH) female patients (N = 16, average age: 46 ± 10.66 years; average education: 13.31 ± 3.8 years; mean duration of illness: 8.56 ± 5.24 years) following an acute stress. Methods: Patients performed a mental stress test followed by three different relaxation sessions presented in a randomized order on three separate occasions. Self-perceived stress, level of relaxation, and heart rate (HR) were monitored during the sessions. Participants' ratings of their experiences were also collected. Results: The results indicated that the three relaxation methods were equally effective in reducing perceived stress induced by acute stress and in lowering HR. However, greater cognitive activation was reported in the two IVR conditions compared with the narrative condition. Conclusions: This is the first study to show evidence of the impact of IVR on a rare population. Despite the lack of significant differences between the two IVR and narrative-alone conditions in physiological and subjective relaxation, more than half of the participants expressed a subjective preference for the virtual experience, especially for the preferred one. © 2026 The Author(s). British Journal of Health Psychology published by John Wiley & Sons Ltd on behalf of British Psychological Society.},

keywords = {adult, controlled study, female, Heart Rate, human, Humans, hypertension, Immersive virtual reality, mental stress, Middle Aged, Narration, pathophysiology, physiology, Pilot Projects, pilot study, procedures, Psychological, psychology, Pulmonary, pulmonary arterial hypertension, pulmonary hypertension, randomized controlled trial, rehabilitation, relaxation, Relaxation Therapy, relaxation training, Stress, Therapy, verbal communication, virtual reality},

pubstate = {published},

tppubtype = {article}

}

@article{damadiFairnessSocialMachines2026a,

title = {Fairness in social machines: a systematic review},

author = {M. S. Damadi and A. Davoust},

url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-105030531101&doi=10.1108%2FJICES-01-2025-0002&partnerID=40&md5=63e2f87ee3852ffe2d49c514e38cba1c},

doi = {10.1108/JICES-01-2025-0002},

issn = {1477996X (ISSN)},

year  = {2026},

date = {2026-01-01},

journal = {Journal of Information, Communication and Ethics in Society},

pages = {1–40},

abstract = {Purpose – The purpose of the paper is to provide a systematic review of biases in social machines to better understand the general problem of fairness in these systems. It aims to identify and categorize phenomena described as biases toward specific demographic groups, frame them normatively as harmful and relate them to established fairness concepts originally defined for algorithmic systems. Design/methodology/approach – The phenomenon of algorithmic bias refers to systematic biases against identifiable demographic groups that occur in automated decisions systems. Such biases have mostly been studied in the context of black-box decision systems built using machine learning (ML). However, similar problems have also been reported in complex socio-technical systems such as Wikipedia and Airbnb, known more generally as social machines, where the observed biases cannot necessarily be attributed to specific automated decision systems. Instead, the biases may emerge as a result of complex processes involving numerous users and a computational infrastructure. To gain a better understanding of fairness in social machines, the authors select a representative sample of social machines from six distinct categories, and systematically review the literature reporting biases in these systems, covering 196 papers. The authors classify the reported bias phenomena, identify the affected demographic groups and relate the phenomena to established notions of harm from algorithmic fairness research. Finally, the authors identify the normative expectations of fairness associated with the different problems and discuss the applicability of existing criteria proposed for ML-driven decision systems. The analysis highlights the conceptual similarity of bias phenomena between algorithmic systems and social machines, allowing for a shared vocabulary to describe and compare phenomena across a broad class of systems. Findings – The paper identifies two key biases in social machines: representational harm, from underrepresentation or biased portrayal of disadvantaged groups, and allocative harm, from unfair decision processes, measurable via metrics like demographic parity. Gender bias is prevalent and easier to detect due to explicit markers, offering insights for identifying other biases. Unique biases arise from user categorizations, creating unintended discrimination linked to protected characteristics. These biases result from complex user interactions, not isolated algorithms. Addressing them requires redesigning social machines, focusing on computational infrastructure and interaction norms, such as visibility settings, to mitigate harmful outcomes. Originality/value – The paper’s originality lies in its systematic review of biases in social machines, offering a novel perspective on fairness in these systems. Unlike prior studies focusing solely on algorithmic fairness, this work examines the broader socio-technical interactions within social machines, identifying biases that emerge from user interactions and design choices. By linking these biases to established fairness concepts like demographic parity and representational harm, the paper bridges the gap between algorithmic fairness and social dynamics. © 2025 Emerald Publishing Limited},

keywords = {Cultural bias, Discrimination, Fairness, Gender bias, Geographic bias, Social machines},

pubstate = {published},

tppubtype = {article}

}

@article{mohamedameyoudMultimodalMalwareClassification2026a,

title = {Multi-modal malware classification with hierarchical consistency and saliency-constrained adversarial training},

author = {S. Mohamed Ameyoud and M. Saïd Allili},

url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-105031186108&doi=10.1016%2Fj.jisa.2026.104429&partnerID=40&md5=2425da4ab40f9043ba4e67d223a1bdd9},

doi = {10.1016/j.jisa.2026.104429},

issn = {22142134 (ISSN)},

year  = {2026},

date = {2026-01-01},

journal = {Journal of Information Security and Applications},

volume = {99},

abstract = {The increasing complexity of malware, including polymorphic, obfuscated, and adversarial variants, continues to outpace the capabilities of detection systems. Here, we introduce a robust multi-modal hierarchical framework that jointly leverages visual and code-level semantics to enhance malware family and type classification. Our architecture fuses convolutional and transformer-based encoders to extract complementary representations from raw malware binaries and decompiled control-flow functions, enabling a rich, cross-modal understanding of malicious behavior. The classification pipeline follows a two-stage hierarchical protocol, where the predicted malware type informs the family-level classification. This enforces ontological consistency between type and family prediction levels. To further bolster robustness against adversarial and obfuscated malware, we integrate a novel adversarial training strategy that generates plausible perturbations guided by attention distributions. Evaluation on multiple large-scale benchmarks including BODMAS, Malimg, Microsoft BIG 2015, and a curated set of from MalwareBazaar, demonstrate that our framework consistently outperforms state-of-the-art baselines, including ResNet, Swin Transformer, and MalBERTv2, across both malware type and family prediction tasks. Notably, our model exhibits outstanding generalization to unpacked, obfuscated, and previously unseen samples, with minimal performance degradation. It achieves accuracy gains of +3-6% over leading methods and exhibits superior resilience under adversarial threat models. These results highlight the effectiveness of hierarchical conditioning, adversarial robustness, and multi-modal fusion in tackling the evolving landscape of malware. The proposed framework thus offers a scalable and generalizable approach for next-generation malware classification in real-world cybersecurity environments. © 2026 Elsevier Ltd.},

keywords = {Adversarial training, Capability of detection, Classification (of information), Convolution, convolutional neural network, Convolutional neural networks, Detection system, Hierarchical consistency, Hierarchical systems, Malware, Malware classification, Malware classifications, Malware families, Malwares, Multi-modal, Multi-modal learning, Semantics, Vision transformer, Vision transformers},

pubstate = {published},

tppubtype = {article}

}

@article{abdollahzadehVisionBasedFrameworkSafe2026a,

title = {A Vision-Based Framework for Safe Landing Zone Mapping of UAVs in Dynamic Environments},

author = {S. Abdollahzadeh and M. S. Allili and A. Boulmerka and J. -F. Lapointe},

url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-105029942397&doi=10.1109%2FOJCS.2026.3663268&partnerID=40&md5=b11484e035458c84b1d3f6780b92c91c},

doi = {10.1109/OJCS.2026.3663268},

issn = {26441268 (ISSN)},

year  = {2026},

date = {2026-01-01},

journal = {IEEE Open Journal of the Computer Society},

volume = {7},

pages = {492–503},

abstract = {Identification safe landing zones (SLZ) for Uncrewed Aerial Vehicles (UAVs) is important to ensure reliable and safe navigation, especially when they are operated in complex and safety-critical environments. However, this is a challenging task due to obstacles and UAV motion. This paper proposes a vision-based framework that maps SLZs in dynamic scenes by integrating several functionalities for analyzing visually static and dynamic aspects of a scene. Static analysis is achieved through context-aware segmentation which divides the image into thematic classes enabling to identify suitable landing surfaces (e.g., roads, grass). For dynamic content analysis, we combine object detection, tracking, and trajectory prediction to determine object occupancy and identify regions free of obstacles. Trajectory prediction is performed through a novel encoder–decoder architecture taking past object positions to predict the most likely future locations. To ensure stable and robust trajectory prediction, we introduce an optimized homography computation using multi-scale image analysis and cumulative updates to compensate UAV motion. We tested our framework on different operational scenarios, including urban and natural scenes with moving objects like vehicles and pedestrians. Obtained results demonstrate its strong performance, and its significant potential for enabling autonomous and safe UAV navigation. © 2020 IEEE.},

keywords = {Aerial vehicle, Air navigation, Aircraft detection, Aircraft landing, Antennas, automatic UAV navigation, Computer vision, Dynamic environments, Forecasting, Homographies, Landing zones, Learning systems, Motion tracking, Object detection, Object recognition, Object Tracking, object trajectory prediction, Robotics, Safe landing, Safe landing zone, safe landing zones (SLZ), Semantic segmentation, Semantics, Trajectories, Trajectory forecasting, Uncrewed aerial vehicles (UAVs), Unmanned aerial vehicle, Unmanned aerial vehicles (UAV)},

pubstate = {published},

tppubtype = {article}

}

@article{gingrasCulturalDifferencesSpatial2026b,

title = {Cultural differences in spatial frequency tunings to faces do not generalize to visual scenes and object stimuli},

author = {F. Gingras and D. Fiset and M. -P. Plouffe-Demers and A. Estéphan and M. N’Guiamba and D. Sun and Y. Zhang and C. Blais},

url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-105025378146&doi=10.3758%2Fs13423-025-02832-0&partnerID=40&md5=43840b8cfa4c2df54e647f03a452f8e5},

doi = {10.3758/s13423-025-02832-0},

issn = {10699384 (ISSN)},

year  = {2026},

date = {2026-01-01},

journal = {Psychonomic Bulletin and Review},

volume = {33},

number = {1},

abstract = {Previous research has identified cultural differences in visual perception, where East Asians focus more on global object structure and display a larger breadth of attention compared with Westerners. East Asians rely on lower spatial frequencies (SFs) compared to Westerners for face recognition, which may be linked to this. Investigating whether such differences extend to other high-level stimulus categories would clarify if SF tuning differences reflect more general or face specific cognitive processes. The present study compared the SF tunings of Canadians and Chinese during object (Exp. 1; N = 50) and scene (Exp. 3; N = 47) categorization. In both experiments, results did not indicate a significant difference between groups. In Experiment 3 (N = 128), we conducted an online replication of Experiment 1 while measuring the SF tunings of the same participants during face perception. Again, no significant difference between the groups was found during object categorization, but the finding that East Asians rely on lower SF than Westerners was replicated. Together, these results suggest that unique mechanisms may underlie the cultural differences in face processing, though alternative explanations, such as the feature consistency of faces, could also account for these findings. © The Psychonomic Society, Inc. 2025.},

keywords = {Adolescent, adult, Asian, Asian People, Canada, Caucasian, China, Cross-Cultural Comparison, Cross-cultural psychology, cultural factor, Depth Perception, Face processing, Facial Recognition, female, human, Humans, male, Object processing, Pattern Recognition, physiology, Scene processing, Space Perception, Spatial frequencies, Visual, visual pattern recognition, Visual Perception, White People, Young Adult},

pubstate = {published},

tppubtype = {article}

}

@article{coteEyesAreCentral2026a,

title = {The eyes are central to face detection: revisiting the foundations of face processing},

author = {L. Côté and J. Lamontagne and A. Bellerose and C. Blais and D. Fiset},

url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-105030389147&doi=10.1016%2Fj.visres.2026.108785&partnerID=40&md5=752aa5d9923ac60539e36118ad41e1e6},

doi = {10.1016/j.visres.2026.108785},

issn = {00426989 (ISSN)},

year  = {2026},

date = {2026-01-01},

journal = {Vision Research},

volume = {243},

abstract = {Face detection feels effortless, yet it requires finely tuned computations to extract socially meaningful signals from the visual stream. Here, we used the Bubbles method to isolate the facial features and spatial frequency information that support face categorization. Across three experiments varying in task demands and visual context, the eye region consistently emerged as the most diagnostic source of information, particularly in high spatial frequencies. This finding held whether participants distinguished faces from noise, from non-face objects, or from real-world categories—suggesting that the eyes serve as an anchor point for categorization across contexts. Strikingly, this diagnostic profile mirrors that found in face identification tasks, implying that detection and recognition may rely on shared perceptual mechanisms rather than sequential, independent processes. This overlap sheds light on longstanding ambiguities in the prosopagnosia literature, indicating that detection impairments found in patients may stem from a broader failure to extract critical eye information. More broadly, our results invite a rethinking of the early stages of face processing, suggesting that detection already involves selective use of diagnostic facial features that supports recognition, emotional decoding, and social perception. © 2026 The Author(s).},

keywords = {adult, article, Black person, Bubbles, Categorization, Caucasian, Detection, emotion assessment, Faces, Facial Recognition, facies, female, human, human experiment, Image analysis, information processing, Information use, male, Noise, normal human, perception, Prosopagnosia, spatial frequency discrimination, task performance, visual discrimination, Young Adult},

pubstate = {published},

tppubtype = {article}

}

@article{gingrasCulturalDifferencesSpatial2026a,

title = {Cultural differences in spatial frequency tunings to faces do not generalize to visual scenes and object stimuli},

author = {F. Gingras and D. Fiset and M. -P. Plouffe-Demers and A. Estéphan and M. N’Guiamba and D. Sun and Y. Zhang and C. Blais},

url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-105025378146&doi=10.3758%2Fs13423-025-02832-0&partnerID=40&md5=43840b8cfa4c2df54e647f03a452f8e5},

doi = {10.3758/s13423-025-02832-0},

issn = {10699384 (ISSN)},

year  = {2026},

date = {2026-01-01},

journal = {Psychonomic Bulletin and Review},

volume = {33},

number = {1},

abstract = {Previous research has identified cultural differences in visual perception, where East Asians focus more on global object structure and display a larger breadth of attention compared with Westerners. East Asians rely on lower spatial frequencies (SFs) compared to Westerners for face recognition, which may be linked to this. Investigating whether such differences extend to other high-level stimulus categories would clarify if SF tuning differences reflect more general or face specific cognitive processes. The present study compared the SF tunings of Canadians and Chinese during object (Exp. 1; N = 50) and scene (Exp. 3; N = 47) categorization. In both experiments, results did not indicate a significant difference between groups. In Experiment 3 (N = 128), we conducted an online replication of Experiment 1 while measuring the SF tunings of the same participants during face perception. Again, no significant difference between the groups was found during object categorization, but the finding that East Asians rely on lower SF than Westerners was replicated. Together, these results suggest that unique mechanisms may underlie the cultural differences in face processing, though alternative explanations, such as the feature consistency of faces, could also account for these findings. © The Psychonomic Society, Inc. 2025.},

keywords = {Adolescent, adult, Asian, Asian People, Canada, Caucasian, China, Cross-Cultural Comparison, Cross-cultural psychology, cultural factor, Depth Perception, Face processing, Facial Recognition, female, human, Humans, male, Object processing, Pattern Recognition, physiology, Scene processing, Space Perception, Spatial frequencies, Visual, visual pattern recognition, Visual Perception, White People, Young Adult},

pubstate = {published},

tppubtype = {article}

}