Search

Nitrogen-rich transition metal nitrides are of great interests due to the unique physical and chemical properties. Here, we perform a detailed structural investigation of ReN8 in the pressure range of 0–200 GPa via particle swarm optimization algorithm and first-principles calculations. Interestingly, four ReN8 phases are firstly revealed to be stable within 200 GPa, N2 dimers are found to be preferred in T-ReN8 at pressure lower than 15.8 GPa and the infinite N∞ chains become dominated in M-, T'- and M'-ReN8 at higher pressure up to 200 GPa. We find that the Young's moduli of M- and T'-ReN8, about 850 GPa along [0.88 0 0.47] and [0 0 1] direction, are comparable to cubic BN, due to the presence of electronic accumulation in the N∞ chains. Furthermore, M- and T'-ReN8 are found to be of high energy density of ~ 1.9 kJ/g (~ 11.0 kJ/cm3). Ultra-incompressibility of ReN8 in special direction due to the high electronic accumulations in the polynitrogen chains. [ABSTRACT FROM AUTHOR]

In this paper, we tackle the problem of how to transfer knowledge from a pre-trained, yet well-performing CNN-based model to train a compact Vision Transformer (ViT)-based model while maintaining the CNN learning capacity? Due to the completely different characteristics of ViT and CNN and the long-existing capacity gap between teacher and student models in Knowledge Distillation (KD), directly transferring the cross-model knowledge is non-trivial. To this end, we subtly leverage the visual and linguistic-compatible feature character of ViT (i. e., student), and its capacity gap with the CNN (i. e., teacher) and propose a novel CNN-to-ViT KD framework, dubbed C2VKD. Importantly, as the teacher's features are heterogeneous to those of the student, we first propose a novel visual-linguistic feature distillation (VLFD) module that explores efficient KD among the aligned visual and linguistic-compatible representations. Moreover, due to the large capacity gap between the teacher and student and the inevitable prediction errors of the teacher, we then propose a pixel-wise decoupled distillation (PDD) module to supervise the student under the combination of labels and teacher's predictions from the decoupled target and non-target classes. Experiments on three semantic segmentation benchmark datasets consistently show that the increment of mIoU of our method is over 200% of the SoTA KD methods. 1 1 Project Page: https://vlislab22.github.io/C2VKD/. • Novel C2VKD framework for learning compact ViT student from CNN teacher. • We present VLFD and PDD modules to distill visual and linguistic features. • C2VKD achieves state-of-the-art performance on three segmentation datasets. [ABSTRACT FROM AUTHOR]

A thorough structural exploration was performed for MgB6 combining the global structure searching method with first-principles calculations. Besides the known Cmcm phase, new phases, i.e. I4/mmm, C2/m-I, C2/m-II and P21/m, were predicted to be stable in the pressure range of 18–100 GPa. Unexpectedly, Cmcm-MgB6 was found to be a semiconductor with an indirect band gap of 0.38 eV with the HSE06 functional, in good agreement with the experimental finding. I4/mmm-MgB6 stabilized above 18 GPa exhibits semimetallic behaviour with a topological node-line near the Fermi level. Consequently, C2/m-I MgB6 with a sandwich structure similar to MgB2 is predicted to be a superconductor with a critical temperature (Tc) of 9.5 K. By analysing the electronic structure, the intriguing semiconductor–semimetal–superconductor transition may be ascribed to the delocalization of more B-p electrons in the boron sublattice. The novel functions uncovered for MgB6 may inspire more efforts to discover materials with intriguing properties. [ABSTRACT FROM AUTHOR]

Benzo(a)Pyrene (BaP) is a well-known environmental carcinogen that poses a significant risk to human health. The pivotal genes and toxicity pathways have been identified as key events to construct the mode of action (MOA) of BaP. In this study, we focused on evaluating the association between genetic variants in BaP-disturbed toxicity pathways and the susceptibility of laryngeal squamous cell carcinoma (LSCC), based on the data of our previous genome-wide association analysis (GWAS). In addition, we investigated the biological roles of these significant polymorphisms by integrating bioinformatic annotation and experimental validation. Our findings revealed that 15 functional polymorphisms in AHR signaling, p53 signaling, NRF2 signaling, TGF-β signaling, STAT3 signaling, and IL-8 signaling pathways were significantly associated with susceptibility to LSCC. Our study provides a novel approach for identifying novel risk genetic loci utilizing GWAS data, and suggests potential targets for early detection of LSCC in the future. [Display omitted] • This study based on the genome-wide association analysis of population. • Integrated the genetic variants in BaP-perturbed toxicity pathways. • 15 functional polymorphisms in six signaling pathways were significantly associated with susceptibility to LSCC. [ABSTRACT FROM AUTHOR]