Search

This article introduces the submission status of the Translation into Low-Resource Languages of Spain task at (WMT 2024) by Huawei Translation Service Center (HW-TSC). We participated in three translation tasks: spanish to aragonese (es-arg), spanish to aranese (es-arn), and spanish to asturian (es-ast). For these three translation tasks, we use training strategies such as multilingual transfer, regularized dropout, forward translation and back translation, labse denoising, transduction ensemble learning and other strategies to neural machine translation (NMT) model based on training deep transformer-big architecture. By using these enhancement strategies, our submission achieved a competitive result in the final evaluation., Comment: 6 pages,wmt24. arXiv admin note: substantial text overlap with arXiv:2409.14842; text overlap with arXiv:2409.14800

Although fusion of information from multiple views of mammograms plays an important role to increase accuracy of breast cancer detection, developing multi-view mammograms-based computer-aided diagnosis (CAD) schemes still faces challenges and no such CAD schemes have been used in clinical practice. To overcome the challenges, we investigate a new approach based on Contrastive Language-Image Pre-training (CLIP), which has sparked interest across various medical imaging tasks. By solving the challenges in (1) effectively adapting the single-view CLIP for multi-view feature fusion and (2) efficiently fine-tuning this parameter-dense model with limited samples and computational resources, we introduce Mammo-CLIP, the first multi-modal framework to process multi-view mammograms and corresponding simple texts. Mammo-CLIP uses an early feature fusion strategy to learn multi-view relationships in four mammograms acquired from the CC and MLO views of the left and right breasts. To enhance learning efficiency, plug-and-play adapters are added into CLIP image and text encoders for fine-tuning parameters and limiting updates to about 1% of the parameters. For framework evaluation, we assembled two datasets retrospectively. The first dataset, comprising 470 malignant and 479 benign cases, was used for few-shot fine-tuning and internal evaluation of the proposed Mammo-CLIP via 5-fold cross-validation. The second dataset, including 60 malignant and 294 benign cases, was used to test generalizability of Mammo-CLIP. Study results show that Mammo-CLIP outperforms the state-of-art cross-view transformer in AUC (0.841 vs. 0.817, 0.837 vs. 0.807) on both datasets. It also surpasses previous two CLIP-based methods by 20.3% and 14.3%. This study highlights the potential of applying the finetuned vision-language models for developing next-generation, image-text-based CAD schemes of breast cancer.

Dielectric environment engineering is an efficient and general approach to manipulating polaritons. Liquids serving as surrounding media of polaritons have been used to shift polariton dispersions and tailor polariton wavefronts. However, those liquid-based methods have so far been limited to their static states, not fully unleashing the promises offered by the mobility of liquids. Here, we propose a microfluidic strategy for polariton manipulation by merging polaritonics with microfluidics. The diffusion of fluids causes gradient refractive indices over microchannels, which breaks the symmetry of polariton dispersions and realizes the non-reciprocal dragging of polaritons. Based on polariton microfluidics, we also design a set of on-chip polaritonic elements to actively shape polaritons, including planar lenses, off-axis lenses, Janus lenses, bends, and splitters. Our strategy expands the toolkit for the manipulation of polaritons at the subwavelength scale and possesses potential in the fields of polariton biochemistry and molecular sensing.

We study the effects of charge-regulated acid/base equilibrium on the swelling of polyelectrolyte gels, by considering a combination of the Poisson-Boltzmann theory and a two-site charge-regulation model based on the Langmuir adsorption isotherm. By exploring the volume change as a function of salt concentration for both nano-gels and micro-gels, we identify conditions where the gel volume exhibits a discontinuous swelling transition. This transition is driven exclusively by the charge-regulation mechanism and is characterized by a closed-loop phase diagram. Our predictions can be tested experimentally for polypeptide gels.

We analyze the translocation process of a spherical vesicle, made of membrane and incompressible fluid, through a hole smaller than the vesicle size, driven by pressure difference $\Delta P$. We show that such a vesicle shows certain universal characteristics which is independent of the details of the membrane elasticity; (i) there is a critical pressure $\Delta P_{\rm c}$ below which no translocation occurs, (ii) $\Delta P_{\rm c}$ decreases to zero as the vesicle radius $R_0$ approaches the hole radius $a$, satisfying the scaling relation $\Delta P_{\rm c} \sim (R_0 - a)^{3/2}$, and (iii) the translocation time $\tau$ diverges as $\Delta P$ decreases to $\Delta P_{\rm c}$, satisfying the scaling relation $\tau \sim (\Delta P -\Delta P_{\rm c})^{-1/2}$.

Deep convolutional neural networks (CNNs) have been widely used in various medical imaging tasks. However, due to the intrinsic locality of convolution operation, CNNs generally cannot model long-range dependencies well, which are important for accurately identifying or mapping corresponding breast lesion features computed from unregistered multiple mammograms. This motivates us to leverage the architecture of Multi-view Vision Transformers to capture long-range relationships of multiple mammograms from the same patient in one examination. For this purpose, we employ local Transformer blocks to separately learn patch relationships within four mammograms acquired from two-view (CC/MLO) of two-side (right/left) breasts. The outputs from different views and sides are concatenated and fed into global Transformer blocks, to jointly learn patch relationships between four images representing two different views of the left and right breasts. To evaluate the proposed model, we retrospectively assembled a dataset involving 949 sets of mammograms, which include 470 malignant cases and 479 normal or benign cases. We trained and evaluated the model using a five-fold cross-validation method. Without any arduous preprocessing steps (e.g., optimal window cropping, chest wall or pectoral muscle removal, two-view image registration, etc.), our four-image (two-view-two-side) Transformer-based model achieves case classification performance with an area under ROC curve (AUC = 0.818), which significantly outperforms AUC = 0.784 achieved by the state-of-the-art multi-view CNNs (p = 0.009). It also outperforms two one-view-two-side models that achieve AUC of 0.724 (CC view) and 0.769 (MLO view), respectively. The study demonstrates the potential of using Transformers to develop high-performing computer-aided diagnosis schemes that combine four mammograms.