Nouveaux Matériaux, Systèmes Intelligents et Entreprises Innovantes

@article{thatte_4055,

title = {The PAH 3.4 micron feature as a tracer of shielding in the Orion Bar and NGC 6240},

author = {Niranjan Thatte and Dimitra Rigopoulou and Fergus R Donnan and Ismael Garcia-Bernete and Miguel Pereira-Santaella and Bruce Draine and Oscar Veenema and Bouthéïna Kerkeni and Almudena Alonso-Herrero and Hermosa Muñoz Laura and Gabriela Speranza},

url = {https://doi.org/10.1093/mnras/staf2047},

year  = {2026},

date = {2026-11-01},

journal = {Monthly Notices Of The Royal Astronomical Society},

volume = {545},

number = {1},

pages = {1-10},

abstract = {We have carried out a detailed analysis of the 3.4 ?m spectral feature arising from Polycyclic Aromatic Hydrocarbons (PAH),

using James Webb Space Telescope archival data. For the first time in an external galaxy (NGC 6240), we have identified two

distinct spectral components of the PAH 3.4 ?m feature: a shorter wavelength component at 3.395 ?m, which we attribute to

short aliphatic chains tightly attached to the aromatic rings of the PAH molecules; and a longer wavelength feature at 3.405 ?m

that arises from longer, more fragile, aliphatic chains that are weakly attached to the parent PAH molecule. These longer chains

are more easily destroyed by far-ultraviolet photons (>5 eV) and PAH thermal emission only occurs where PAH molecules

are shielded from more energetic photons by dense molecular gas. We see a very strong correlation in the morphology of the

PAH 3.395 ?m feature with the PAH 3.3 ?m emission, the latter arising from robust aromatic PAH molecules. We also see an

equally strong correlation between the PAH 3.405 ?m morphology and the warm molecular gas, as traced by H2 vibrational

lines. We show that the flux ratio PAH 3.395/PAH 3.405 < 0.3 corresponds strongly to regions where the PAH molecules are

shielded by dense molecular gas, so that only modestly energetic UV photons penetrate to excite the PAHs. Our work shows that

PAH 3.405 ?m and PAH 3.395 ?m emission features can provide robust diagnostics of the physical conditions of the interstellar

medium in external galaxies, and can be used to quantify the energies of the photon field penetrating molecular clouds.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{xiang_4433,

title = {An adaptive-weighted multi-level feature evidence and redundancy penalty method},

author = {Feng Xiang and Chunxiao Xing and Ye Ai and Zhiqiang Wang},

url = {https://doi.org/10.1016/j.engappai.2026.115245},

year  = {2026},

date = {2026-09-01},

journal = {Engineering Applications Of Artificial Intelligence},

volume = {180},

pages = {115245},

abstract = {Aiming to improve quality indicator prediction for the tobacco drying process in a cigarette factory, this study proposes an adaptive-weighted multi-level feature evidence and redundancy-penalty framework (F-XMF). To alleviate the limitations of isolated feature evaluation in complex industrial data, F-XMF jointly considers the direct contribution of individual features, the contribution of feature interactions, and the redundant dependency among candidate features. First, eXtreme Gradient Boosting (XGBoost) is used to estimate the importance of individual features and obtain an initial main-effect feature subset. Meanwhile, an Attentional Factorization Machine (AFM) is employed to identify high-contribution second-order interactions and derive interaction-related evidence. On this basis, conditional mutual information (CMI) is introduced to characterize redundant dependency among candidate features, and a Differential Redundancy Penalty Mechanism is further developed to impose differentiated penalties on highly overlapping features while preserving informative variables with meaningful interaction utility. Finally, a validation-based adaptive weight fusion strategy is used to integrate the main-effect subset and the interaction-related subset, yielding a compact and informative final feature subset. Experimental results on real tobacco-processing data show that F-XMF consistently reduces Mean Absolute Error (MAE) in ablation comparisons and achieves superior performance against multiple representative baseline methods. Mechanism validation further demonstrates that the proposed framework can effectively refine the candidate subset through interaction enhancement and redundancy suppression, showing good interpretability and practical value for outlet moisture content prediction in the tobacco drying process.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{xiao_4539,

title = {Enhancing semi-supervised multi-view graph convolutional networks via supervised contrastive learning and self-training},

author = {Huaiyuan Xiao and Fadi Dornaika and Jinan Charafeddine and Jingjun Bi},

url = {https://doi.org/10.1016/j.knosys.2026.116500},

year  = {2026},

date = {2026-09-01},

journal = {Knowledge-Based Systems},

volume = {349},

pages = {116500},

abstract = {The advent of graph convolutional network (GCN)-based multi-view learning provides a powerful framework for integrating structural information from heterogeneous views, enabling effective modeling of complex multi-view data. However, existing methods often fail to fully exploit the complementary information across views, leading to suboptimal feature representations and limited performance. To address this, we propose MV-SupGCN, a semi-supervised GCN model that integrates several complementary components with clear motivations and mutual reinforcement. First, to better capture discriminative features and improve model generalization, we design a joint loss function that combines Cross-Entropy loss with Supervised Contrastive loss, encouraging the model to simultaneously minimize intra-class variance and maximize inter-class separability in the latent space. Second, recognizing the instability and incompleteness of single graph construction methods, we combine both KNN-based and semi-supervised graph construction approaches on each view, thereby enhancing the robustness of the data structure representation and reducing generalization error. Third, to effectively utilize abundant unlabeled data and enhance semantic alignment across multiple views, we propose a unified framework that integrates contrastive learning in order to enforce consistency among multi-view embeddings and capture meaningful inter-view relationships, together with pseudo-labeling, which provides additional supervision applied to both the cross-entropy and contrastive loss functions to enhance model generalization. Extensive experiments demonstrate that MV-SupGCN consistently surpasses state-of-the-art methods across multiple benchmarks, validating the effectiveness of our integrated approach.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{ghorbel_4371,

title = {Mechanical behavior of apatite-alumina coatings deposited by high-velocity oxy-fuel spray (SHVOF): Prosthesis application},

author = {Halima Ghorbel and Radouane Akrache and Jinan Charafeddine and Awatef Guidara},

url = {https://doi.org/10.1016/j.nxmate.2026.102076},

year  = {2026},

date = {2026-07-01},

journal = {Next Materials},

volume = {12},

pages = {102076},

abstract = {Alumina-fluorapatite (Al?O?-Fap) composite coatings deposited on 316?L stainless steel by Suspension High-Velocity Oxy-Fuel (SHVOF) spraying were investigated as candidate surfaces for prosthetic applications. Compared with conventional atmospheric plasma spraying (APS), SHVOF processing is known to produce denser coatings with improved adhesion, which is of interest for load-bearing biomedical interfaces. In this study, low fluorapatite contents (0-1?wt%) were incorporated into an alumina matrix to explore their influence on coating mechanical behavior. The coatings were characterized in terms of surface roughness, adhesion strength, and elastic properties. The results indicate that the addition of fluorapatite modifies the coating morphology and leads to an increase in surface roughness (Ra from ?1.7 to ?3.1 ?m), while maintaining adhesion values within the range reported for thermally sprayed bioceramic coatings. An increase in elastic modulus was also observed, from 59.7?GPa for pure alumina to 98.3?GPa for Al?O?-Fap (1?wt%), corresponding to an increase of approximately 64%, indicating a marked modification of the coating mechanical response. In parallel, a three-dimensional finite element (FE) model of a cylindrical 316?L implant inserted into cortical bone was developed using Abaqus/CAE to simulate the early stage of implantation. The model was used to evaluate von Mises stresses and contact pressures at the bone-implant interface. The simulations predict von Mises stresses of approximately 48?MPa and contact pressures ranging from 25 to 55?MPa during insertion, values consistent with those reported for cortical bone under similar conditions. However, these results depend on simplifying assumptions, including linear elastic and isotropic bone behavior and the absence of an explicit coating layer. Overall, this combined experimental-numerical approach provides quantitative insight into the effect of low fluorapatite additions on the mechanical performance of SHVOF-deposited alumina coatings. The results suggest that such coatings may contribute to improved mechanical compatibility at the implant interface, although further biological and in vivo investigations are required to confirm their clinical relevance.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{lebon_4378,

title = {Machinability of complex concentrated alloys for bio-implants: Roughness analysis after machining},

author = {Nicolas Lebon and Bernardin Mawussi},

url = {https://www.sciencedirect.com/science/article/pii/S2352492826010238?via%3Dihub},

year  = {2026},

date = {2026-06-01},

journal = {Materials Today Communications},

volume = {54},

number = {115634},

pages = {1-15},

abstract = {TiNbZrMoTa-based HEAs produced by laser powder bed fusion (L-PBF) are promising candidates for biomedical implants, but their post-processing machinability remains insufficiently understood. This study investigates the influence of additive manufacturing conditions and machining strategies on the cutting forces and surface integrity of 3 TiNbZrMoTa-based HEAs manufactured under different volumetric energy densities. Five machining configurations were evaluated at different feedrates.

Cutting force measurements revealed a significant influence of both alloy grade and machining strategy. Under carbide cutting conditions, mean cutting forces increased from 28.8±1.2 N for VED550 to 57.3±1.4 N for VED350 and 61.1±2.7 N for VED88. Abrasive machining reduced machining loads up to 68% compared with cutting tools. Feedrate increased cutting forces but had little influence on surface roughness.

Cutting tool consistently produced the smoothest surfaces, with Ra/Sa values below 0.30 µm. The VED88 alloy machined with a cutting tool exhibited the best overall performance, achieving Ra = 0.118±0.018 µm, Rz = 0.765±0.144 µm, Sa = 0.143±0.010 µm, and Sz = 2.365±0.533 µm. One-way ANOVA demonstrated a highly significant effect of machining configuration on roughness parameters (p < 0.001), whereas ultrasonic assistance mainly improved process stability by reducing Sz variability from 7.31 to 2.45 µm. The results identify VED88 machined with a carbide cutting tool as the optimal compromise between machinability, surface integrity, and process repeatability for biomedical implant applications. These results provide guidance for the post processing of additive manufactured HEAs and support their integration into machined biomedical implant where surface integrity is critical for functional performance.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{zalan_4453,

title = {Metaverse Alchemy: Connecting NFTs, Crypto-Tokens, and Social Media in Web3 Platform Multinationals' Success},

author = {Tatiana ZALAN and Efstathios Polyzos and Elissar Toufaily},

url = {https://doi.org/10.1016/j.lrp.2026.102647},

year  = {2026},

date = {2026-06-01},

journal = {Long Range Planning},

pages = {102647},

abstract = {This paper focuses on the emerging internet phenomenon of Web3 platforms which are a new generation of multinationals. Using two popular metaverse platforms, Decentraland and The Sandbox, we examine the nexus between the platform NFTs (non-fungible tokens), the cryptocurrency (i.e., the native token) and the social media discussions around the platforms. We combine token price data with information from OpenSea and X (Twitter) and implement wavelet coherence analysis to examine the time-varying relationship between these three pillars of performance. Our analysis suggests that it is the NFT market (and not the cryptocurrency market) that drives the performance of the decentralized metaverse platforms. Social media discussions generate buzz and sentiment, acting as a catalyst of success. We contribute to shedding light on ?the big question? in international business (IB) research (Peng, 2004) and on the issue of user adoption in the novel context of Web3 PMNCs (platform multinational corporations). The practical implication of our research is that Web3 PMNC builders should design business models that provide utility to NFTs.},

keywords = {},

pubstate = {online},

tppubtype = {article}

}

@article{khafaga_4498,

title = {An Integrated NB-IoT/5G Architecture Leveraging Fabry-Perot Optical Sensors for Enhanced Energy Efficiency and Coverage},

author = {Ali Salah Mahdi Khafaga and Dhurgham Ali Saeed ALRasheed and Wafaa M. R. Shakir and Jinan Charafeddine},

url = {https://doi.org/10.48084/etasr.18034},

year  = {2026},

date = {2026-06-01},

journal = {Engineering, Technology & Applied Science Research},

volume = {16},

number = {3},

pages = {35129-35135},

abstract = {In the current era of rapid development in communications technology, fifth-generation (5G) networks, next-generation networks, and the Internet of Things (IoT), many challenges have appeared, the most important being energy consumption and coverage. Although 5G networks offer high data speeds and low latency, the use of millimeter waves limits their range. However, the Narrowband Internet of Things  (NB-IoT) is characterized by wide coverage and low energy consumption, but does not have the bandwidth to accommodate all applications that require high data transfer speeds. This study presents an integrated communication model that uses Fabry-Perot optical sensors and combines NB-IoT and 5G network technologies. Fabry-Perot sensors are integrated into a network infrastructure to feed the Software-Defined Network / Network Function Virtualization (SDN/NFV) controller, which dynamically manages resources, leading to a quantitative improvement in key simulation results (38.67% energy savings and improved spectrum efficiency). This model improves both energy efficiency and coverage and is scalable and effective for next-generation networks, especially in IoT networks. The model was simulated using MATLAB, demonstrating the scalable performance and effective integration of optical sensing with network control.},

keywords = {},

pubstate = {online},

tppubtype = {article}

}

@article{cheng_4501,

title = {Dynamic demand forecasting in bike-sharing systems: a multidimensional spatiotemporal network},

author = {Quan Cheng and Xin Li and Xiaoli SU and Zhe Yuan},

url = {https://doi.org/10.1080/00207543.2026.2683107},

year  = {2026},

date = {2026-06-01},

journal = {International Journal Of Production Research},

pages = {1-31},

abstract = {Accurate demand forecasting is critical for optimising resource allocation and operational efficiency in bike-sharing systems (BSS). However, this remains challenging owing to the complex nonlinear characteristics of large-scale data, irregular spatiotemporal patterns, and external environmental uncertainties. To address these issues, this study proposes a multidimensional spatiotemporal fusion framework that combines Domain-Adaptive Density Clustering (DADC) and an Attention Temporal Graph Convolutional Network (A3T-GCN). First, the DADC algorithm addresses data sparsity by adaptively clustering stations based on spatial density variations, thereby effectively capturing dynamic distribution patterns. Second, a weighted directed graph is constructed incorporating station utility and full-load-rate functions to eliminate low-performance nodes. A dynamic graph sequence model is then developed to characterise the temporal evolution of spatial dependencies and enhance peak-period responsiveness. The A3T-GCN architecture integrates station-specific attributes with environmental factors through attention mechanisms, enabling the focused analysis of critical demand influencers while processing multidimensional spatiotemporal dependencies. Experimental validation using Citi Bike datasets demonstrates superior performance over benchmark models, reducing prediction errors by 11.54% (15?min) and 12.33% (60?min) through improved short-term accuracy and long-term stability. The framework effectively addresses the uncertainty in demand forecasting, providing methodological support for sustainable BSS operations through enhanced spatiotemporal pattern recognition and environmental adaptability.},

keywords = {},

pubstate = {online},

tppubtype = {article}

}

@article{gregory_4507,

title = {The role of venture?financed startups in innovation for US agriculture},

author = {Graff Gregory and Charles de Grazia and E. Rada Nicholas},

url = {https://doi.org/10.1002/ajae.70083},

year  = {2026},

date = {2026-06-01},

journal = {American Journal Of Agricultural Economics},

abstract = {Technology diffusion and spillovers are key drivers of both innovation and economic growth. This paper examines the effects of obtaining initial intellectual property rights on international knowledge flows through new technological entrants, an important yet understudied function of the IP system. We propose a simple theoretical framework which examines exclusivity and proximity effects as countervailing forces acting on the broader diffusion of foreign-developed technology within a host country. Our study provides causal evidence that a foreign technological entrant's initial patent grant in the host country (the United States) increases the likelihood of international knowledge flows to local U.S. firms and other patenting entities (7.1 percentage points and 29.4 percent, respectively), demonstrating the importance of patent rights in facilitating international knowledge flows beyond the disclosure mechanism. These effects are at least partly driven by sustained presence, cross-country differences in search costs, and voluntary technology sales, but the evidence on inventor mobility is mixed. Overall, initial intellectual property rights appear to mitigate impediments to cross-border knowledge flows in this setting, and the benefits of doing so outweigh the frictions arising from the exclusionary nature of patents.},

keywords = {},

pubstate = {online},

tppubtype = {article}

}

@article{lebon_4518,

title = {A computer-aided manufacturing tool for analyzing the tool-workpiece contact during the micro grinding of dental prostheses},

author = {Nicolas Lebon and Laurent Tapie},

url = {https://doi.org/10.21203/rs.3.rs-9860168/v1},

year  = {2026},

date = {2026-06-01},

journal = {International Journal Of Advanced Manufacturing Technology},

abstract = {Dental CAD/CAM manufacturing predominantly relies on three-axis milling strategies, which impose geometric and kinematic constraints that influence machining conditions and prosthesis quality. This study investigates tool-workpiece contact configurations and effective grinding speed variations during the micro-grinding of dental prostheses. Two abrasive burs were geometrically characterized and modeled using an APT-based analytical formalism. The resulting models enabled the determination of local contact conditions and effective grinding speeds on freeform crown geometries.

The methodology was implemented on a clinical dental crown model to evaluate the spatial distribution of tool engagement, grinding speed, and machining accessibility. Results showed that, regardless of tool geometry, machining is performed almost exclusively by the hemispherical tip of the bur, while conical and cylindrical sections contribute marginally. Consequently, effective grinding speed exhibits strong spatial variability driven mainly by local surface orientation. The lowest speeds occur at cusp tips, whereas higher values are observed along peripheral regions, suggesting significant variations in surface integrity, roughness, and tool wear across the prosthesis.

A curvature-based accessibility analysis revealed that 40,1% of the crown surface is inaccessible under three-axis machining constraints, particularly within anatomically important grooves and fossae. Because these regions contribute to food guidance, fragmentation, and occlusal contact distribution during mastication, their incomplete reproduction may compromise the functional anatomy of the restoration and reduce masticatory efficiency.

These findings demonstrate the strong coupling between prosthesis geometry, tool design, and machining kinematics, highlighting the need for advanced manufacturing strategies, such as multi-axis machining, to improve accessibility, preserve functional morphology, and enhance prosthesis performance.},

keywords = {},

pubstate = {online},

tppubtype = {article}

}

@article{wang_4532,

title = {Optimizing parallel machine location and scheduling with sequence-dependent setup times via logic-based benders decomposition},

author = {Yun Wang and Peng Wu and Zhe Yuan and Yeming Gong and Ajay Kumar},

url = {https://doi.org/10.1016/j.ejor.2026.06.027},

year  = {2026},

date = {2026-06-01},

journal = {European Journal Of Operational Research},

abstract = {ow that (i) The proposed FLBBD algorithm can solve a significantly greater number of instances to optimality compared with the off-the-shelf MILP solver CPLEX and BD incorporated within CPLEX; (ii) The FLBBD algorithm finds feasible solutions for 100% instances whereas CPLEX and BD solve only 81.17% and 57.65% instances; (iii) The FLBBD algorithm attains a smaller average optimality gap of 4.89%, in contrast to the 59.10% and 26.09% average optimality gap observed for CPLEX and BD; and (iv) The derived valid inequalities reduces 6.67% computational time and 62.37% optimality gap compared with CPLEX.},

keywords = {},

pubstate = {online},

tppubtype = {article}

}

@article{maroun_4537,

title = {A Global-Context Vision Transformer for Facial Age Estimation},

author = {Ghaby Maroun and Salah Eddine Bekhouche and Jinan Charafeddine and Fadi Dornaika},

url = {https://doi.org/10.1007/s12559-026-10605-7},

year  = {2026},

date = {2026-06-01},

journal = {Cognitive Computation},

volume = {18},

number = {1},

pages = {72},

abstract = {Estimating a person's age from a single facial image remains challenging because ageing cues are subtle, non-linear, and

easily confounded by identity, pose, illumination, and demographic factors. In this work, we investigate the suitability of

the Global Context Vision Transformer (GC-ViT), originally introduced for general vision tasks, as a pure transformer

backbone for facial age estimation. We extend GC-ViT to scalar age regression through a lightweight MLP head and

a task-specific training strategy, with the emphasis placed on a targeted reformulation of the architecture for facial age

estimation and on its comprehensive empirical validation in this setting. GC-ViT is well suited to age estimation because

it combines short-range window self-attention with long-range global attention, enabling a single hierarchical transformer

to capture both fine texture cues (e.g., wrinkles) and global facial proportions. To improve robustness under limited data,

we introduce an age-aware region replacement augmentation that replaces semantically defined facial regions with age-

proximal donor content, or with self-flipped target content, and blends them using a feathered mask together with area-

proportional soft label interpolation in donor mode. Training uses an ablation-validated composite regression loss (MAE

with a secondary stabilising term, MSE or SmoothL1 depending on the dataset). We evaluate on MORPH II, IMDB-Clean,

AFAD, and UTKFace, reporting MAE together with Pearson correlation and cumulative scores (CS@5/CS@10). Our GC-

ViT achieves an MAE of 2.41 on MORPH II, 3.11 on AFAD, and 4.26 on UTKFace (the latter two obtained by fine-tuning

from IMDB-Clean pretraining), while also being assessed under zero-shot cross-dataset transfer to quantify domain shift

and out-of-domain generalization. Overall, the results show that an adapted local-global attention transformer, trained

with targeted augmentation and a stable regression objective, can serve as a strong transformer-based baseline for facial

age estimation without auxiliary CNN branches or test-time post-processing.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{boukhari_4538,

title = {H-ARN: A holo-attentive relational network for holistic facial beauty prediction via distribution learning},

author = {Djamel Boukhari and Fadi Dornaika and Jinan Charafeddine},

url = {https://doi.org/10.1007/s13735-026-00401-2},

year  = {2026},

date = {2026-06-01},

journal = {International Journal Of Multimedia Information Retrieval},

volume = {15},

number = {3},

pages = {20},

abstract = {Facial Beauty Prediction (FBP) remains a significant challenge in computer vision, primarily due to the inherent subjectivity

of human perception and the difficulty in modeling the holistic harmony of facial features. Existing deep learning models,

typically based on Convolutional Neural Networks (CNNs), often regress a deterministic score and struggle to surpass

a performance plateau. In this work, we propose the Holo-Attentive Relational Network (H-ARN), a novel architecture

designed to overcome these fundamental limitations. H-ARN makes two primary contributions: (1) It pioneers the use of

distribution learning by predicting the mean and variance of beauty scores, employing a Gaussian Negative Log-Likelihood

loss to robustly model the subjectivity and ambiguity in training data. (2) It introduces a relational self-attention module

atop a powerful Vision Transformer (ViT) backbone, explicitly learning the complex, long-range relationships between facial

components that define aesthetic harmony. Evaluated on the challenging SCUT-FBP5500 benchmark, our proposed H-ARN

achieves a new state-of-the-art Pearson Correlation of 0.9333, significantly outperforming previous methods. The code will

be made publicly available upon publication https://github.com/DjameleddineBoukhari/H-ARN.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{pignot_4069,

title = {Ethical design: a post-Lacanian ethnography of a serious healthcare game},

author = {Edouard Pignot and Jean-Etienne Joullié and Anthony Gould},

url = {https://doi.org/10.1108/JOE-06-2025-0084},

year  = {2026},

date = {2026-05-01},

journal = {Journal of Organizational Ethnography},

volume = {15},

number = {1},

pages = {1-24},

abstract = {Purpose

This ethnographic case study explores how designers of a serious healthcare game internalise ethical challenges.



Design/methodology/approach

The analysis uses Jason Glynos' post-Lacanian theory of ethics with its underpinning notions of fantasy, identification and detachment to interpret how designers wrestle with infantilisation (characterised here as overidentification with fantasy) insofar as it affects elderly people.



Findings

The research finds that game design ethically benefits when designers experience fantasy traversal.



Originality/value

Games as a means of promoting healthcare and lifestyle outcomes raise ethical conundrums. To address these concerns, researchers have generally adopted either a normative or a virtue ethics perspective of game design. In so doing, they have neglected how designers actually experience creating a game and, in particular, make decisions about the infantilisation that the process sometimes entails. The present study seeks to remedy such oversight.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{wang_4202,

title = {SCS-SupCon: Sigmoid-based common and style supervised contrastive learning with adaptive decision boundaries},

author = {Bin Wang and Fadi Dornaika and Jinan Charafeddine},

url = {http://dx.doi.org/10.1016/j.eswa.2026.131308},

year  = {2026},

date = {2026-05-01},

journal = {Expert Systems With Applications},

volume = {311},

pages = {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 con-

trastive 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, explic-

itly parameterized by learnable temperature and bias terms. This design places greater emphasis on critical dis-

criminative 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 percent-

age 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.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{el_magri_4432,

title = {Investigating the effect of printing parameters on the porosity and tensile behavior of 3D-printed PEEK in Z-direction},

author = {Anouar El Magri and Zakariae Bouchkara and Sébastien Vaudreuil and Hamidreza Vanaei},

url = {https://doi.org/10.1007/s40964-026-01718-4},

year  = {2026},

date = {2026-05-01},

journal = {Progress in Additive Manufacturing},

volume = {11},

number = {5},

pages = {1-19},

abstract = {Additive manufacturing (AM) is an innovative technology widely used in several sectors, including aerospace, electronics, automotive and medical. One of the processes used is deposition of molten filaments using a material. In this study, Poly (ether ether ketone) was used as the base material for additive manufacturing via the fused deposition modeling (FFF) process. The main objective of this study was to investigate how variations in nozzle temperature, layer thickness, and printing speed influence mechanical properties and porosity, with a focus on the Z-direction, which is particularly prone to weak interlayer bonding. By systematically analyzing these effects, the study provides new insights into optimizing FDM process parameters to improve interlayer adhesion and the overall structural integrity of vertically printed components. In addition, we carried out characterization using and X-ray diffraction analyses dynamic mechanical analysis (DMA) and scanning electron microscopy (SEM). Using Response Surface Methodology (RSM), optimal printing conditions were predicted as 405 °C nozzle temperature, 60 mm/s printing speed, and 0.1 mm layer thickness, yielding components with a Young's modulus of 3302 MPa, tensile strength of 72 MPa, strain at break of 3.25%, and a porosity rate of 4.70%, closely matching the experimental observations. These results may indicate that the optimized printing conditions predicted by RSM can reliably produce PEEK components with balanced mechanical performance and low porosity.},

keywords = {},

pubstate = {online},

tppubtype = {article}

}

@article{pairolero_4505,

title = {Automated routing, patent examiner specialization, and quasi-random assignment at the USPTO},

author = {A. Nicholas Pairolero and Charles de Grazia},

url = {https://doi.org/10.1016/j.respol.2026.105460},

year  = {2026},

date = {2026-05-01},

journal = {Research Policy},

volume = {55},

number = {4},

pages = {105460},

abstract = {In October 2020, the United States Patent and Trademark Office (USPTO) introduced an automated process for routing new patent applications, in part to increase examiner specialization. In this research note, we first assess whether the USPTO achieved this objective using a difference-in-differences (DiD) research design on a sample of applications assigned to examiners between 2018 and 2021. The new routing system increased technical overlap between examiner work histories and newly assigned cases by 57 percent, indicating a substantial increase in examiner specialization across the agency. In addition, we show that automated routing ? largely unknown in the academic literature ? effectively eliminated quasi-random assignment of applications to examiners. Given the importance of quasi-random assignment for empirical research, we discuss our findings within this context and potential impact on future research.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{arya_3724,

title = {AI-Enhanced Competency Transfer Hubs: A Conceptual Framework for University-Industry Engagement and Knowledge Sharing},

author = {Vikas Arya and Ankita Saraf and Nikolai Chichkanov and Armando Papa and Marco Romano},

url = {https://doi.org/10.1007/s10961-025-10233-7},

year  = {2026},

date = {2026-04-01},

journal = {Journal Of Technology Transfer},

volume = {51},

pages = {682-712},

abstract = {This paper introduces a framework for AI-driven competency transfer hubs, designed to facilitate effective knowledge exchange and collaboration between universities and industries. These hubs leverage artificial intelligence technologies like machine learning and natural language processing to enhance the efficiency and effectiveness of information flows between academic institutions and industry partners, optimizing the whole knowledge-sharing process. Using the TCM-ADO framework the paper consolidates existing perspectives and offers practical suggestions on how to incorporate AI technologies into competency hubs. The discussion further delves into outlining key layers of such hubs including AI-powered knowledge extraction and enrichment, knowledge customization, adaptive project management as well as collaboration outcome enhancement and feedback optimization. A set of key elements for AI-enhanced competency transfer hubs was also developed and presented including the issues of technical alignment, advanced AI integration as well as value aspects. The study wraps up by exploring key areas of application in the establishment of AI-enhanced competency transfer hubs and their wider societal significance.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{wang_4305,

title = {Multi-Agent Transformer Learning for Moving Target Positioning and Tracking in Complex Environments Using UAV Swarms},

author = {Haowen Wang and Junyu Wei and Ni ZHU and Zongqing Zhao and Zhuoyuan Wu and Shiqi Li and Yuyang Xiao and Jiangyi Qin and Zhiqiang Wang},

url = {https://ieeexplore.ieee.org/document/11478047/keywords#keywords},

year  = {2026},

date = {2026-04-01},

journal = {Ieee Transactions On Intelligent Transportation Systems},

pages = {1-17},

abstract = {This paper presents a novel approach that integrates Time Difference of Arrival (TDOA)-based positioning with Multi-Agent Reinforcement Learning (MARL) to enhance the performance of Unmanned Aerial Vehicle (UAV) swarms in tracking and locating a moving ground target within complex environments. To ensure optimal target assignment, we employ the Hungarian algorithm, enabling each UAV to track a designated target and maintain a spatial formation that minimizes the geometric dilution of precision (GDOP). The coordinated behavior of the UAV swarm is then formulated as a partially observable Markov decision process (POMDP), where a Multi-Agent Transformer (MAT) model, built upon the multi-agent advantage decomposition theorem, is employed to map UAV observations to coordinated actions effectively. Finally, the UAVs adjust their trajectories based on the generated actions to improve positioning accuracy, avoid obstacles, and conserve energy. Through extensive simulations, the proposed approach demonstrates its effectiveness in continuously and accurately tracking a moving target, even in dynamic and challenging environments.},

keywords = {},

pubstate = {online},

tppubtype = {article}

}

@article{de_grazia_4506,

title = {Foreign technological entry, intellectual property rights, and technology diffusion: evidence from patent filings at the USPTO},

author = {Charles de Grazia and Herman Peter and Hwansung Ju},

url = {https://doi.org/10.1007/s10961-026-10336-9},

year  = {2026},

date = {2026-04-01},

journal = {Journal Of Technology Transfer},

abstract = {Technology diffusion and spillovers are key drivers of both innovation and economic growth. This paper examines the effects of obtaining initial intellectual property rights on international knowledge flows through new technological entrants, an important yet understudied function of the IP system. We propose a simple theoretical framework which examines exclusivity and proximity effects as countervailing forces acting on the broader diffusion of foreign-developed technology within a host country. Our study provides causal evidence that a foreign technological entrant's initial patent grant in the host country (the United States) increases the likelihood of international knowledge flows to local U.S. firms and other patenting entities (7.1 percentage points and 29.4 percent, respectively), demonstrating the importance of patent rights in facilitating international knowledge flows beyond the disclosure mechanism. These effects are at least partly driven by sustained presence, cross-country differences in search costs, and voluntary technology sales, but the evidence on inventor mobility is mixed. Overall, initial intellectual property rights appear to mitigate impediments to cross-border knowledge flows in this setting, and the benefits of doing so outweigh the frictions arising from the exclusionary nature of patents.},

keywords = {},

pubstate = {online},

tppubtype = {article}

}

@article{du_4028,

title = {Sequentiality and Simultaneity in R&D Collaboration With Multiple Partner Types: Complexity and Managerial Resource Constraints in R&D Projects},

author = {Jingshu Du and Rene Belderbos and Bart Leten and Wim Vanhaverbeke},

url = {https://doi.org/10.1111/radm.70037},

year  = {2026},

date = {2026-03-01},

journal = {R & D Management},

volume = {56},

number = {2},

pages = {410-423},

abstract = {R&D collaborations involving multiple types of partners have the potential to generate impactful innovations but also pose significant coordination challenges. Taking a coordination perspective, this study develops and tests a theoretical framework

on the optimal temporal coordination of partner relationships in such collaborations. We argue that the effective coordination of multiple types of collaboration partners in R&D projects requires a balanced approach, combining episodes of simultaneous involvement of all partners with periods of sequential engagement with individual partners. Moreover, we propose that the relative emphasis on simultaneous versus sequential involvement depends on project-specific factors, including the complexity of technology development and the availability of managerial resources. Our analysis of 454 R&D projects undertaken by a

major European electronics firm, in collaboration with diverse partner types (science-based partners such as universities and

research institutions, versus market-based partners such as suppliers and customers), provides robust empirical support for these hypotheses.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{bi_4203,

title = {A Re-Node self-training approach for deep graph-based semi-supervised classification on multi-view image data},

author = {Jingjun Bi and Fadi Dornaika and Jinan Charafeddine},

url = {https://doi.org/10.1016/j.inffus.2025.103887},

year  = {2026},

date = {2026-03-01},

journal = {Information Fusion},

volume = {127},

number = {Part C},

pages = {103887},

abstract = {Recently, graph-based semi-supervised learning and pseudo-labeling have gained attention due to their effectiveness in reducing the need for extensive data annotations. Pseudo-labeling uses predictions from unlabeled data to improve model training, while graph-based methods are characterized by processing data represented as graphs. However, the lack of clear graph structures in images combined with the complexity of multi-view data limits the efficiency of traditional and existing techniques. Moreover, the integration of graph structures in multi-view data is still a challenge. In this paper, we propose Re-node Self-taught Graph-based Semi-supervised Learning for Multi-view Data (RSGSLM). Our method addresses these challenges by (i) combining linear feature transformation and multi-view graph fusion within a Graph Convolutional Network (GCN) framework, (ii) dynamically incorporating pseudo-labels into the GCN's loss function to improve classification in multi-view data, and (iii) correcting topological imbalances by adjusting the weights of labeled samples near class boundaries. Additionally, (iv) we introduce an unsupervised smoothing loss applicable to all samples. This combination optimizes performance while maintaining computational efficiency.

Experimental results on multi-view benchmark image datasets demonstrate that RSGSLM surpasses existing semi-supervised learning approaches in multi-view context},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{taieb_4286,

title = {Online live streaming interaction: Development and validation of a measurement scale},

author = {Basma Taieb and Jean-Éric Pelet and Xuemei Sun},

url = {https://doi.org/10.1108/IJRDM-01-2025-0044},

year  = {2026},

date = {2026-03-01},

journal = {International Journal Of Retail & Distribution Management},

pages = {1-16},

abstract = {Purpose

Live streaming commerce is an emerging shopping channel that facilitates real-time, two-way interaction not only between consumers and vendors, but also among consumers. This study conceptualizes Online Live Streaming Interaction (OLSI) as a multidimensional construct, grounded in social presence theory, and develops a measurement tool that captures the unique dynamics of interactions in live streaming environments.

Design/methodology/approach

The study follows a rigorous, four-step process for scale development and validation. Inductive and deductive methods were employed for item generation. Quantitative data were then collected from 394 Chinese participants and 159 French participants, ensuring cross-cultural relevance. The data were analyzed using exploratory and confirmatory factor analyses to assess the reliability and validity of the scale.

Findings

The results establish the OLSI scale as a multidimensional construct encompassing six key dimensions: online discussion engagement, streamer-viewer interactions, preference for online social interactions, communication effectiveness for streamers, online interaction propensity, and perceived value of the online community. 

Originality

This study provides a comprehensive framework for understanding the unique interaction dynamics within live streaming commerce and offers a robust measurement tool for future research. 

Research limitations/implications

This study's primary limitation is its limited exploration of nomological validity, future research should investigate the relationships between the OLSI dimensions and behavioural outcomes (e.g., purchase intention, satisfaction, loyalty). 

Practical implications 

The findings provide practical insights for managers and enable them to develop content and engagement strategies that align with audience preferences; this will lead to both an optimization of communication and enhancement of viewer experiences.},

keywords = {},

pubstate = {online},

tppubtype = {article}

}

@article{ben_elechi_4057,

title = {Analysis of cambering, twist, and opening sections for springback prediction},

author = {Slim Ben Elechi and Mohamed Guerich},

url = {https://doi.org/10.1556/606.2025.01443},

year  = {2026},

date = {2026-02-01},

journal = {Pollack Periodica},

abstract = {The accurate prediction of springback in sheet metal forming remains a significant challenge in industrial manufacturing, given its impact on dimensional accuracy and product quality. The present study investigates the effects of both process and material parameters on springback, focusing on geometric deviations like cambering, twist, and opening in the final part. A comprehensive analysis encompassed three coefficients of friction, four constitutive models, and three yield functions. The deep drawing of an automotive side sill closing was selected as a case study representative of the subject. The findings indicate that the coefficient of friction exerts a predominant influence on all three springback indicators - cambering, twist, and opening. Conversely, the selection of constitutive model has a comparatively negligible effect, particularly on twist prediction.},

note = {Volume/Issue:

    Accepted Manuscript / Online First},

keywords = {},

pubstate = {online},

tppubtype = {article}

}

@article{mohamed_taha_4077,

title = {Optimization of the Performance of Bimorph Piezoelectric Energy Harvesters with Honeycomb Metamaterials Using Artificial Neural Networks and NSGA-II},

author = {Mhiri Mohamed Taha and Walid Larbi and Mnaouar Chouchane and Mohamed Guerich},

url = {https://doi.org/10.13052/ejcm2642-2085.34346},

year  = {2026},

date = {2026-02-01},

journal = {European Journal of Computational Mechanics},

volume = {34},

number = {3&4},

pages = {325-362},

abstract = {Honeycomb-based metamaterials have recently attracted considerable interest for their potential in energy-harvesting applications. In this paper, we focus on two types of bimorph harvesters incorporating honeycomb substrates with positive and negative Poisson's ratios (PPR and NPR). The substrates are coupled with piezoceramic (PZT) layers to enable electromechanical conversion. To improve energy-harvesting efficiency, we propose an optimization framework that integrates finite element (FE) simulations for data generation, a pretrained neural network for rapid performance prediction and the NSGA-II evolutionary algorithm for multiobjective optimization. The proposed strategy enables the identification of optimal geometric parameters of the honeycomb cells, particularly for cantilever resonators where the fundamental vibration modes are critical and must be tuned to specific eigenfrequencies for engineering applications. The optimization is applied to an initially unoptimized bimorph harvester with a honeycomb substrate. The simulation results indicate that an optimized harvester with an NPR metamaterial substrate can increase the power-to-mass ratio at a resonance frequency of 160 Hz by approximately 14.72% compared to the unoptimized honeycomb harvester. Moreover, compared to a harvester with a solid substrate, the improvement reaches 5%. Furthermore, a life cycle analysis was conducted, showing that the honeycomb substrate can significantly increase the operational lifetime of the piezoceramic layer. This improvement arises because the honeycomb substrate reduces the equivalent stress on the active layer. At high acceleration levels, the predicted operational lifetime is enhanced by more than tenfold compared to the solid substrate, since the PZT material is brittle and lacks flexibility. This improvement is notable because it surpasses the typical lifespan of both standard button batteries and rechargeable batteries highlighting a strong potential of this design for future applications.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{gayatri_4207,

title = {Fast interlayer energy transfer from the lower bandgap MoS2 to the higher bandgap WS2},

author = {Gayatri Gayatri and Mehdi Arfaoui and Debashish Das and Tomasz Kazimierczuk and Sabrine Ayari and Natalia Zawadzka and Takashi Taniguchi and Kenji Watanabe and Adam Babi?ski and Saroj K Nayak and Maciej R Molas and Arka Karmakar},

url = {https://doi.org/10.1038/s41699-026-00661-w},

year  = {2026},

date = {2026-02-01},

journal = {Npj 2D Materials And Applications},

volume = {10},

number = {25},

abstract = {Energy transfer (ET) is a dipole-dipole interaction, mediated by the virtual photon. Traditionally, ET happens from the higher (donor) to lower bandgap (acceptor) material. However, in some rare instances, ET can happen from the lower-to-higher bandgap material, depending on the strong overlap between the acceptor photoluminescence (PL) and the donor absorption spectra. In this work, we report an ET process from the lower bandgap MoS2 to the higher bandgap WS2, due to a near ?resonant' overlap between the MoS2 B and WS2 A excitonic levels. Changing the MoS2 bandgap from direct-to-indirect by increasing the layer number results in a reduced ET rate, evidenced by the quenching of the WS2 PL emission. Our work shows at 300?K, the ET timescale of ~33?fs is faster than the reported thermalization of the MoS2 excitonic intervalley scattering (K???K') time and competing with the ultrafast charge transfer timescale. Thus, allowing us to open a new direction in understanding the competing inter/intralayer processes.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{ketata_4268,

title = {Performance Enhancement of Bonded Aluminum Composite Panels through Control of Polyethylene Core Thickness},

author = {HAJER KETATA and Naila Hfaiedh and Hatem Ksibi},

url = {https://doi.org/10.1021/acsomega.5c12525},

year  = {2026},

date = {2026-02-01},

journal = {Acs Omega},

volume = {1},

abstract = {This study analyzes the global mechanical behavior and performance optimization of Aluminum Composite Panels (ACP) with a polyethylene (PE) core. Conventional multilayer ACP designs often lead to excessive material use and high weight. The effect of PE core thickness on mechanical performance was investigated to reduce the number of layers while maintaining structural integrity. Experimental tensile tests and finite element (FEM) simulations using ABAQUS/Explicit were conducted to evaluate the elastic and plastic deformation of ACP up to failure. The combined effects of core thickness and the number of sandwich layers were systematically analyzed. Results show that the apparent Young's modulus (Eapp) decreases with increasing core thickness and number of layers. Adjusting the PE core thickness allows a single ACP to replace multiple bonded panels while maintaining equivalent load-bearing capacity. A single ACP with a 10 mm PE core can replace three 4 mm bonded panels. This approach reduces aluminum consumption, structural weight, and material usage while promoting the use of recyclable polymer cores. The findings highlight the relevance of the apparent modulus as a key design parameter and confirm the potential of ACPs as lightweight, high-performance, and sustainable materials for industrial applications.},

keywords = {},

pubstate = {online},

tppubtype = {article}

}

@article{chu_3793,

title = {A general framework and an individual dominance realization model for efficient fixed cost allocation based on DEA},

author = {Junfei Chu and Yanhua Dong and Zhe Yuan and Weijiao Wang},

url = {https://doi.org/10.1016/j.omega.2025.103382},

year  = {2026},

date = {2026-01-01},

journal = {Omega-International Journal Of Management Science},

volume = {138},

pages = {103382},

abstract = {Data Envelopment Analysis (DEA) has become a powerful tool for addressing fixed cost allocation (FCA) problems. Among the numerous DEA-based FCA studies, a significant category seeks to make all Decision-making Units (DMUs) efficient under a set of common weights after cost allocation, falling under the common-weights FCA framework. However, the classical DEA efficiency evaluation mechanism does not require all DMUs to be evaluated based on a set of common weights. Instead, a DMU is considered efficient if it can choose its favorable weights to achieve an efficiency score of one while ensuring that the efficiencies of other DMUs do not exceed one. Motivated by this, we develop a novel and general efficient FCA framework that allocates the total fixed cost by allowing each DMU to select its preferred weights to be efficient. We prove that the FCA possible set under the traditional common-weights framework is a subset of that under this new framework. Then, an Assurance Region (AR) constraint on fixed costs is introduced to avoid extreme allocation schemes. Moreover, an individual dominance realization strategy and a FCA model based on it are proposed within this framework, aimed at minimizing each DMU's cross-efficiencies from the perspective of all other DMUs while ensuring its self-assessed efficiency remains at one. Additionally, an algorithm is introduced to guarantee the uniqueness of the FCA scheme. A numerical example and a case study from the Shanghai Automotive Industry Corporation are used to validate the effectiveness of this approach and compare it with other classic methods.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{micica_3947,

title = {Terahertz Electronic and Spin Currents in Wafer?Scale Van der Waals Bi2Se3/WSe2 Heterostructures and Polymorphs},

author = {Martin Mi?ica and Adrien Wright and Robson Ferreira and Sabrine Ayari and Enzo Rongione and Mario Oliveira Ribeiro and Sajid Husain and Rahul Sharma and Thibaud Denneulin and Rafak.E Dunin?Borkowski and Mangeney Juliette and Jérôme Tignon and R. Lebrun and Hanoko Okuno and Olivier Boulle and Alain Marty, and Frédéric Bonell and Francesca Carosella and Henri Ja?rés and R. Ferreira and Jean?Marie George and Jamet Matthieu and Dhillon Sukhdeep},

url = {http://dx.doi.org/10.1002/adma.202506031},

year  = {2026},

date = {2026-01-01},

journal = {Advanced Materials},

volume = {38},

number = {1},

pages = {e06031},

abstract = {Van der Waals heterostructures and polymorphs have promised the realizationof arti?cial materials with multiple physical phenomena such as giant opticalnonlinearities, spin-to-charge interconversion in spintronics and topologicalcarrier protection, through an in?nitely diverse set of 2D quantum materialsand their stacking order in a single layered device. However, their exploitationfor the terahertz range has been limited with most investigations based aroundthe optical domain, owing to the use of exfoliated material that inherently limitsboth the dimensions of the materials and the scalability for applications. Here,the combination of terahertz electronic and spin currents is demonstratedthrough the realization of large area complex crystalline heterostructuresof topological insulators, transition metal dichalcogenides (TMDs)and ferromagnets. This is demonstrated through down-conversion of opticalbeams into coherent terahertz currents, where the terahertz phase permitsthe decoupling of the physical phenomena, and bringing novel functionalitiesbeyond those achievable in simple homostructures. In particular, the role ofdi?erent TMD polymorphs (stacking orders - 1T?, 2H, and 3R) is shown with thesimple change of one atomic monolayer of the material stack entirely changingthe terahertz responses - both electrical and magnetic - of the arti?cialmaterial. This allows to highlight ultrafast phenomena that combine both theelectronic- and spin-based processes in these structures. The importance ofthe crystal symmetry on the magnetic properties through proximity e?ects isfurther demonstrated, showing a nonlinearity of magnetic origin as a result ofthe 1T? polymorph. As well as control of the terahertz currents through di?erentpolymorphs, this scalable integration of a set of highly diverse materialsestablishes a platform for next-generation 2D heterostructures that integratephotonic, electronic, and spintronic properties into device architectures},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{micica_4074,

title = {Terahertz Electronic and Spin Currents in Wafer?Scale Van der Waals Bi                    2                    Se                    3                    /WSe                    2                    Heterostructures and Polymorphs},

author = {Martin Mi?ica and Adrien Wright and Sylvain Massabeau and Sabrine Ayari and Enzo Rongione and Mario Oliveira Ribeiro and Sajid Husain and Rahul Sharma and Thibaud Denneulin and Rafak.E Dunin?Borkowski and Juliette Mangeney and Jérôme Tignon and Romain Lebrun and Hanoko Okuno and Olivier Boulle and Alain Marty, and Frédéric Bonell and Francesca Carosella and Henri Ja?rés and Robson Ferreira and Jean?Marie George and Jamet Matthieu and Dhillon Sukhdeep},

url = {https://doi.org/10.1002/adma.202506031},

year  = {2026},

date = {2026-01-01},

journal = {Advanced Materials},

volume = {38},

number = {1},

pages = {e06031},

abstract = {Van der Waals heterostructures and polymorphs have promised the realization of artificial materials with multiple physical phenomena such as giant optical nonlinearities, spin-to-charge interconversion in spintronics and topological carrier protection, through an infinitely diverse set of 2D quantum materials and their stacking order in a single layered device. However, their exploitation for the terahertz range has been limited with most investigations based around the optical domain, owing to the use of exfoliated material that inherently limits both the dimensions of the materials and the scalability for applications. Here, the combination of terahertz electronic and spin currents is demonstrated through the realization of large area complex crystalline heterostructures of topological insulators, transition metal dichalcogenides (TMDs) and ferromagnets. This is demonstrated through down-conversion of optical beams into coherent terahertz currents, where the terahertz phase permits the decoupling of the physical phenomena, and bringing novel functionalities beyond those achievable in simple homostructures. In particular, the role of different TMD polymorphs (stacking orders - 1T?, 2H, and 3R) is shown with the simple change of one atomic monolayer of the material stack entirely changing the terahertz responses - both electrical and magnetic - of the artificial material. This allows to highlight ultrafast phenomena that combine both the electronic- and spin-based processes in these structures. The importance of the crystal symmetry on the magnetic properties through proximity effects is further demonstrated, showing a nonlinearity of magnetic origin as a result of the 1T? polymorph. As well as control of the terahertz currents through different polymorphs, this scalable integration of a set of highly diverse materials establishes a platform for next-generation 2D heterostructures that integrate photonic, electronic, and spintronic properties into device architectures.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{wang_4139,

title = {Bi-Objective optimisation for synchronising replenishment and storage assignment: achieving energy efficiency and blocking avoidance in forward-reserve warehousing},

author = {Lin Wang and Ning Zhang and Sirui Wang and Xuerui Wang and Zhe Yuan},

url = {https://doi.org/10.1080/00207543.2026.2618574},

year  = {2026},

date = {2026-01-01},

journal = {International Journal Of Production Research},

abstract = {The forward-reserve warehouse is an integrated robotic storage and picking system, which combines

a reserve area equipped with an automatic storage/retrieval system and a forward area supported

by a robotic mobile fulfilment system. Blocking in FRWs leads to energy dissipation and inefficiency,

strongly affected by storage assignment and replenishment timeliness. However, existing research

rarely focuses on systematically quantifying the benefits of replenishment and storage assignment

synchronous optimisation with blocking avoidance. To address this gap, we formulate a new bi-

objective model to improve both energy efficiency and timeliness, explicitly developing two tailored

mathematical evaluation models for energy consumption and blocking. We design a novel ENSGA-II

algorithm and integrate it with an LSTM network to solve the problem. Experimental results show

that the ENSGA-II achieves superior distribution and faster convergence, as measured by HV/GD/IGD

across all scales. We also conduct a comprehensive sensitivity analysis and discuss the results in

detail. The findings reveal that a medium number of picking stations ensures the best balance

between energy efficiency and timeliness objectives. The FRW achieves optimal performance when

the station ratio between the forward and reserve areas remains in a medium range (e.g., 4/20 to

12/8). Additionally, SKU-type diversity in the forward area significantly affects system costs. Both

low and high diversity reduce costs, while medium diversity (e.g., 0.5) results in the least benefits.

In particular, small-scale warehouses benefit more from higher SKU-type diversity},

keywords = {},

pubstate = {online},

tppubtype = {article}

}

@article{taieb_4201,

title = {Le processus d'intention d'achat en ligne à l'ère du live streaming commerce?: rôle central de l'expérience de flow},

author = {Basma Taieb and Jean-Éric Pelet},

url = {https://doi.org/10.1177/07673701251408},

year  = {2026},

date = {2026-01-01},

journal = {Recherche et Application en Marketing},

volume = {41},

number = {1},

pages = {96-123},

abstract = {Cette recherche propose un modèle intégré expliquant comment la confiance du consommateur et l'expérience de flow s'articulent de manière complémentaire dans le processus d'intention d'achat lors du live streaming commerce (LSC). Ce modèle, fondé sur les théories de l'échange social et du flow, introduit deux routes principales : la route cognitive établit le lien entre les caractéristiques du streamer et la confiance envers le produit et la route affective associe les caractéristiques du LSC à l'expérience de flow. Une troisième route qualifiée de mixte représente la relation entre la confiance et l'expérience de flow. Cette route mixte illustre les mécanismes séquentiels conduisant à l'intention d'achat en ligne. Dans un premier temps, une étude qualitative a permis de consolider la pertinence des concepts intégrés dans le modèle de recherche. Par la suite, une étude quantitative réalisée en France auprès de 555 répondants a permis de tester le modèle proposé. Les résultats révèlent la médiation totale de la confiance concernant l'effet des caractéristiques du streamer sur l'intention d'achat. Ils montrent également la médiation totale de l'expérience de flow. La confiance représente ainsi une base cognitive nécessaire mais insuffisante pour accroître l'intention d'achat en ligne. Ce n'est que par son effet médiatisé via l'expérience de flow que cette intention s'intensifie. La mise en évidence de l'articulation entre la confiance (cognition) et le flow (affect) éclaire les mécanismes sous-jacents à l'intention d'achat en ligne et ouvre de nouvelles perspectives de recherche.},

keywords = {},

pubstate = {published},

tppubtype = {article}

}

@article{boersma_4206,

title = {Integrated Strategies for Union Legitimacy (Re)Building: A Process Model},

author = {Martijn Boersma and Emmanuel Josserand},

url = {https://doi.org/10.1111/irel.70021},

year  = {2026},

date = {2026-01-01},

journal = {Industrial Relations},

abstract = {This study presents a longitudinal analysis examining the process by which the Transport Workers Union (re)established its legitimacy through an integrated strategy encompassing comprehensive campaigns in road transport, aviation, and the gig economy in Australia. We reveal how unions can gain pragmatic legitimacy through power recomposition, moral legitimacy through moral framing, and cognitive legitimacy through the shaping of assumptions. We reveal the systematic transfer of proven strategies to different parts of the industry, adapting to specific conditions and audiences. Our findings provide a framework for unions to revitalize their influence within a dynamic industrial relations landscape.},

keywords = {},

pubstate = {online},

tppubtype = {article}

}