Table summarization can be of great help, which generates a concise and informative overview of a table to assist users to understand the table easily and unambiguously. A high-quality summary needs ...to have two desirable properties: presenting notable entities in the table and achieving broad coverage and high diversity on domains. However, notability and domain are often neglected in table summarization. Thus in this paper, we present a framework of domain-aware table summarization that is able to: (1) identify notable entities using a popularity-sensitive notability evaluation algorithm, (2) find core domains with a measurement of domain centrality, (3) and output the final high-quality summary using a three-phase clustering based algorithm. The experimental results show that our summarization method outperforms state-of-the-art methods by 9.62%, 2.78% and 6.77% on metrics coverage, diversity, and notability, respectively. We also conduct a user study to demonstrate that people can improve the accuracy of understanding tables by 17% with the help of our summarization technique.
Speech enhancement in the time domain is becoming increasingly popular in recent years, due to its capability to jointly enhance both the magnitude and the phase of speech. In this work, we propose a ...dense convolutional network (DCN) with self-attention for speech enhancement in the time domain. DCN is an encoder and decoder based architecture with skip connections. Each layer in the encoder and the decoder comprises a dense block and an attention module. Dense blocks and attention modules help in feature extraction using a combination of feature reuse, increased network depth, and maximum context aggregation. Furthermore, we reveal previously unknown problems with a loss based on the spectral magnitude of enhanced speech. To alleviate these problems, we propose a novel loss based on magnitudes of enhanced speech and a predicted noise. Even though the proposed loss is based on magnitudes only, a constraint imposed by noise prediction ensures that the loss enhances both magnitude and phase. Experimental results demonstrate that DCN trained with the proposed loss substantially outperforms other state-of-the-art approaches to causal and non-causal speech enhancement.
• Cells are continuously exposed to chemical signals that they must discriminate between and respond to appropriately. In embryophytes, the leucine-rich repeat receptor-like kinases (LRR-RLKs) are ...signal receptors critical in development and defense. LRR-RLKs have diversified to hundreds of genes in many plant genomes. Although intensively studied, a well-resolved LRR-RLK gene tree has remained elusive.
• To resolve the LRR-RLK gene tree, we developed an improved gene discovery method based on iterative hidden Markov model searching and phylogenetic inference. We used this method to infer complete gene trees for each of the LRR-RLK subclades and reconstructed the deepest nodes of the full gene family.
• We discovered that the LRR-RLK gene family is even larger than previously thought, and that protein domain gains and losses are prevalent. These structural modifications, some of which likely predate embryophyte diversification, led to misclassification of some LRR-RLK variants as members of other gene families. Our work corrects this misclassification.
• Our results reveal ongoing structural evolution generating novel LRR-RLK genes. These new genes are raw material for the diversification of signaling in development and defense. Our methods also enable phylogenetic reconstruction in any large gene family.
This paper proposes a dense network composed of an improved Transformer network, which successfully restores low-light images to high-quality normal-light images, alleviating issues such as low ...brightness, high noise, and missing critical information in low-light images. The entire network architecture is based on the improved Transformer network and builds a dense network with a combination of long and short connections. While retaining the self-attention mechanism of the Transformer network, it achieves multi-level fusion and utilization of shallow and deep features, providing the network with rich image features and enabling the restoration of low-light images to high-quality normal-light images. Additionally, a spatial-domain and frequency-domain combined loss function is designed, considering both pixel-level and frequency domain losses, effectively constraining the image restoration process and avoiding spectral biases. Lastly, a multi-scale hybrid gate feedforward network is designed to replace the traditional feedforward network in the Transformer, facilitating feature selection and forward propagation. These designs effectively enhance the richness of meaningful image features, alleviate spectral biases, and improve the visual quality of low-light images. Experimental results demonstrate the superiority of our method over state-of-the-art networks on various typical image enhancement datasets. Taking the most commonly used low-light dataset LOLv1 as an example, our method achieves improvements of 1.3% and 3.07% in PSNR and SSIM, respectively, compared to the best-performing network, showing favorable qualitative and quantitative evaluation results. The proposed method effectively addresses the issue of insufficiently realistic results in low-light image restoration, providing a reliable reference for practical applications.
In hyperspectral image (HSI) classification, the challenge of the small-sample-size problem persists as a significant obstacle due to the high cost of labeling samples. To effectively train models ...with a limited sample set, the application of a transfer learning approach called cross-scene HSI classification is considered a viable solution to address this problem. In cross-scene HSI classification, a source scene with sufficient labeled samples is leveraged to assist in classifying a target scene that lacks labeled samples. Considering that real HSIs may be captured by different sensors, we propose a novel heterogeneous transfer learning algorithm called dual-stream discriminative attention network (DSDAN) to address the task of cross-scene HSI classification. The DSDAN predominantly comprises three pivotal modules: 1) a dual-stream lightweight hybrid CNN (DSLHC) incorporates both the source stream and the target stream and is applied to extract alignment spatial-spectral features from heterogeneous data; 2) a discriminative attention block (DAB) is created to address the domain shift between two scenes. Following the DSLHC, the DAB assigns discriminative attention weights to the source features, facilitating a closer alignment of features from two scenes; and 3) a specially designed cross-domain loss (CDL) is designed to drive intraclass samples from two scenes to become more consistent, while interclass samples from two scenes become more distinct, thereby further mitigating domain shift. By combining DSLHC, DAB, and CDL, the complete DSDAN model is established. The effectiveness of DSDAN is validated using three real cross-scene HSI datasets.
Gene duplication is a key mechanism for the adaptive evolution and neofunctionalization of gene families. Large multigene families often exhibit complex evolutionary histories as a result of frequent ...gene duplication acting in concordance with positive selection pressures. Alterations in the domain structure of genes, causing changes in the molecular scaffold of proteins, can also result in a complex evolutionary history and has been observed in functionally diverse multigene toxin families. Here, we investigate the role alterations in domain structure have on the tempo of evolution and neofunctionalization of multigene families using the snake venom metalloproteinases (SVMPs) as a model system. Our results reveal that the evolutionary history of viperid (Serpentes: Viperidae) SVMPs is repeatedly punctuated by domain loss, with the single loss of the cysteine-rich domain, facilitating the formation of P-II class SVMPs, occurring prior to the convergent loss of the disintegrin domain to form multiple P-I SVMP structures. Notably, the majority of phylogenetic branches where domain loss was inferred to have occurred exhibited highly significant evidence of positive selection in surface-exposed amino acid residues, resulting in the neofunctionalization of P-II and P-I SVMP classes. These results provide a valuable insight into the mechanisms by which complex gene families evolve and detail how the loss of domain structures can catalyze the accelerated evolution of novel gene paralogues. The ensuing generation of differing molecular scaffolds encoded by the same multigene family facilitates gene neofunctionalization while presenting an evolutionary advantage through the retention of multiple genes capable of encoding functionally distinct proteins.
Difficulty in obtaining adequate abdominal wall closure due to loss of the abdominal domain is a frequent complication of multivisceral, isolated intestinal transplantation and in some cases of liver ...transplantation. Various methods for primary closure have been proposed, including the use of synthetic and biological meshes, as well as full‐thickness abdominal wall and non‐vascularized rectus fascia grafts. We describe a novel technique for abdominal wall procurement in which the graft is perfused synchronously with the abdominal organs and can be transplanted as a full‐thickness wall or as a non‐vascularized rectus fascia graft. We performed six transplants of non‐vascularized rectus fascia in three intestinal transplants, one multivisceral transplant, and two liver transplants. The size of the covered abdominal wall defects ranged from 17 cm × 7 cm to 25 cm × 20 cm. Only one patient developed graft infection secondary to enterocutaneous fistula requiring surgical correction and removal of the fascia graft. This patient, as well as two other patients, died due to sepsis. Our procurement technique allows removal of the rectus fascia graft to cover the abdominal wall defect, providing a feasible solution for treatment of abdominal wall defects in recipients after abdominal organ transplantation.
Neural-Network-Based Cross-Channel Intra Prediction Li, Yue; Yi, Yan; Liu, Dong ...
ACM transactions on multimedia computing communications and applications,
08/2021, Letnik:
17, Številka:
3
Journal Article
Recenzirano
To reduce the redundancy among different color channels, e.g., YUV, previous methods usually adopt a linear model that tends to be oversimple for complex image content. We propose a ...neural-network-based method for cross-channel prediction in intra frame coding. The proposed network utilizes twofold cues, i.e., the neighboring reconstructed samples with all channels, and the co-located reconstructed samples with partial channels. Specifically, for YUV video coding, the neighboring samples with YUV are processed by several fully connected layers; the co-located samples with Y are processed by convolutional layers; and the proposed network fuses the twofold cues. We observe that the integration of twofold information is crucial to the performance of intra prediction of the chroma components. We have designed the network architecture to achieve a good balance between compression performance and computational efficiency. Moreover, we propose a transform domain loss for the training of the network. The transform domain loss helps obtain more compact representations of residues in the transform domain, leading to higher compression efficiency. The proposed method is plugged into HEVC and VVC test models to evaluate its effectiveness. Experimental results show that our method provides more accurate cross-channel intra prediction compared with previous methods. On top of HEVC, our method achieves on average 1.3%, 5.4%, and 3.8% BD-rate reductions for Y, Cb, and Cr on common test sequences, and on average 3.8%, 11.3%, and 9.0% BD-rate reductions for Y, Cb, and Cr on ultra-high-definition test sequences. On top of VVC, our method achieves on average 0.5%, 1.7%, and 1.3% BD-rate reductions for Y, Cb, and Cr on common test sequences.
Even though Norwegian is the predominant language in almost all sectors of society in Norway, there has been an increasing tendency in the university sector in the recent years to introduce English ...as a medium of instruction, particularly at the post-graduate level. Using English has for some years been politically encouraged as part of internationalisation efforts, while the questions of who, where and when have largely been left up to the individual university departments and staff. This paper presents a case study of one such university department, which conducts all their teaching through the medium of English. The study asks the questions: In which ways is English being used? Has the department's English-only policy resulted in English only being used, or are Norwegian and other languages also used in certain circumstances, regardless of the policy? Why did this particular university department choose to make English its official language of instruction? The paper relates Fishman's domain theory to code-switching theory, as defined by Fishman, Auer and Heller. It further discusses whether domain- and code-switching theory is compatible with Bourdieu's theory of language and symbolic power, and Anderson's theory on imagined communities, and whether the combined application of these theories may shed light on the linguistic situation in academia.
The last two decades have seen the rise of an academic and political debate in Denmark about the growing influence of Global English in many contexts of society. Some measures were taken to limit the ...consequences of such influence, especially the loss of domain in higher education, research, and business. However, Danes are usually considered, at home and abroad, to be extremely proficient in English, to the point of being deemed bilingual, and the attitude towards English is generally positive in Denmark. The purpose of the survey in this paper was to determine the extent of use of English in four social practices, as well as the attitude towards learning English, language death, and bilingualism. The results found that Danes generally do not perceive their language to be at risk, at least not in the majority of contexts. They consider English an important language to learn for study and work, but Danish still seems to be in a dominant position in everyday life.
The last two decades have seen the rise of an academic and political debate in Denmark about the growing influence of Global English in many contexts of society. Some measures were taken to limit the consequences of such influence, especially the loss of domain in higher education, research, and business. However, Danes are usually considered, at home and abroad, to be extremely proficient in English, to the point of being deemed bilingual, and the attitude towards English is generally positive in Denmark. The purpose of the survey in this paper was to determine the extent of use of English in four social practices, as well as the attitude towards learning English, language death, and bilingualism. The results found that Danes generally do not perceive their language to be at risk, at least not in the majority of contexts. They consider English an important language to learn for study and work, but Danish still seems to be in a dominant position in everyday life.