Search

Climate warming is a pressing global issue with substantial impacts on soil health and function. However, the influence of environmental context on the responses of soil microorganisms to warming remains largely elusive, particularly in alpine ecosystems. This study examined the responses of the soil microbiome to in situ experimental warming across three elevations (3850 m, 4100 m, and 4250 m) in the meadow of Gongga Mountain, eastern Tibetan Plateau. Our findings demonstrate that soil microbial diversity is highly resilient to warming, with significant impacts observed only at specific elevations. Furthermore, the influence of warming on the composition of the soil microbial community is also elevation-dependent, underscoring the importance of local environmental context in shaping microbial evolution in alpine soils under climate warming. Notably, we identified soil moisture at 3850 m and carbon-to-nitrogen ratio at 4250 m as indirect predictors regulating the responses of microbial diversity to warming at specific elevations. These findings underscore the paramount importance of considering pre-existing environmental conditions in predicting the response of alpine soil microbiomes to climate warming. Our study provides novel insights into the intricate interactions between climate warming, soil microbiome, and environmental context in alpine ecosystems, illuminating the complex mechanisms governing soil microbial ecology in these fragile and sensitive environments.

Existing self-supervised methods in natural language processing (NLP), especially hierarchical text classification (HTC), mainly focus on self-supervised contrastive learning, extremely relying on human-designed augmentation rules to generate contrastive samples, which can potentially corrupt or distort the original information. In this paper, we tend to investigate the feasibility of a contrastive learning scheme in which the semantic and syntactic information inherent in the input sample is adequately reserved in the contrastive samples and fused during the learning process. Specifically, we propose an information lossless contrastive learning strategy for HTC, namely \textbf{H}ierarchy-aware \textbf{I}nformation \textbf{L}ossless contrastive \textbf{L}earning (HILL), which consists of a text encoder representing the input document, and a structure encoder directly generating the positive sample. The structure encoder takes the document embedding as input, extracts the essential syntactic information inherent in the label hierarchy with the principle of structural entropy minimization, and injects the syntactic information into the text representation via hierarchical representation learning. Experiments on three common datasets are conducted to verify the superiority of HILL., Comment: Accepted by NAACL 2024

In the field of urban planning, general-purpose large language models often struggle to meet the specific needs of planners. Tasks like generating urban planning texts, retrieving related information, and evaluating planning documents pose unique challenges. To enhance the efficiency of urban professionals and overcome these obstacles, we introduce PlanGPT, the first specialized Large Language Model tailored for urban and spatial planning. Developed through collaborative efforts with institutions like the Chinese Academy of Urban Planning, PlanGPT leverages a customized local database retrieval framework, domain-specific fine-tuning of base models, and advanced tooling capabilities. Empirical tests demonstrate that PlanGPT has achieved advanced performance, delivering responses of superior quality precisely tailored to the intricacies of urban planning.

Learning time-series representations for discriminative tasks has been a long-standing challenge. Current pre-training methods are limited in either unidirectional next-token prediction or randomly masked token prediction. We propose a novel architecture called Bidirectional Timely Generative Pre-trained Transformer (BiTimelyGPT), which pre-trains on time-series data by both next-token and previous-token predictions in alternating transformer layers. This pre-training task preserves original distribution and data shapes of the time-series. Additionally, the full-rank forward and backward attention matrices exhibit more expressive representation capabilities. Using biosignal data, BiTimelyGPT demonstrates superior performance in predicting neurological functionality, disease diagnosis, and physiological signs. By visualizing the attention heatmap, we observe that the pre-trained BiTimelyGPT can identify discriminative segments from time-series sequences, even more so after fine-tuning on the task.

We present a novel prompt-based personalized federated learning (pFL) method to address data heterogeneity and privacy concerns in traditional medical visual question answering (VQA) methods. Specifically, we regard medical datasets from different organs as clients and use pFL to train personalized transformer-based VQA models for each client. To address the high computational complexity of client-to-client communication in previous pFL methods, we propose a succinct information sharing system by introducing prompts that are small learnable parameters. In addition, the proposed method introduces a reliability parameter to prevent the negative effects of low performance and irrelevant clients. Finally, extensive evaluations on various heterogeneous medical datasets attest to the effectiveness of our proposed method., Comment: Accept by ICASSP2024

Recent advancements on Large Language Models (LLMs) enable AI Agents to automatically generate and execute multi-step plans to solve complex tasks. However, since LLM's content generation process is hardly controllable, current LLM-based agents frequently generate invalid or non-executable plans, which jeopardizes the performance of the generated plans and corrupts users' trust in LLM-based agents. In response, this paper proposes a novel ``Formal-LLM'' framework for LLM-based agents by integrating the expressiveness of natural language and the precision of formal language. Specifically, the framework allows human users to express their requirements or constraints for the planning process as an automaton. A stack-based LLM plan generation process is then conducted under the supervision of the automaton to ensure that the generated plan satisfies the constraints, making the planning process controllable. We conduct experiments on both benchmark tasks and practical real-life tasks, and our framework achieves over 50% overall performance increase, which validates the feasibility and effectiveness of employing Formal-LLM to guide the plan generation of agents, preventing the agents from generating invalid and unsuccessful plans. Further, more controllable LLM-based agents can facilitate the broader utilization of LLM in application scenarios where high validity of planning is essential. The work is open-sourced at https://github.com/agiresearch/Formal-LLM., Comment: 21 pages, 6 figures; comments and suggestions are welcome

The semantically interactive radiance field has always been an appealing task for its potential to facilitate user-friendly and automated real-world 3D scene understanding applications. However, it is a challenging task to achieve high quality, efficiency and zero-shot ability at the same time with semantics in radiance fields. In this work, we present FastLGS, an approach that supports real-time open-vocabulary query within 3D Gaussian Splatting (3DGS) under high resolution. We propose the semantic feature grid to save multi-view CLIP features which are extracted based on Segment Anything Model (SAM) masks, and map the grids to low dimensional features for semantic field training through 3DGS. Once trained, we can restore pixel-aligned CLIP embeddings through feature grids from rendered features for open-vocabulary queries. Comparisons with other state-of-the-art methods prove that FastLGS can achieve the first place performance concerning both speed and accuracy, where FastLGS is 98x faster than LERF and 4x faster than LangSplat. Meanwhile, experiments show that FastLGS is adaptive and compatible with many downstream tasks, such as 3D segmentation and 3D object inpainting, which can be easily applied to other 3D manipulation systems.

Code completion models have made significant progress in recent years. Recently, repository-level code completion has drawn more attention in modern software development, and several baseline methods and benchmarks have been proposed. However, existing repository-level code completion methods often fall short of fully using the extensive context of a project repository, such as the intricacies of relevant files and class hierarchies. Besides, the existing benchmarks usually focus on limited code completion scenarios, which cannot reflect the repository-level code completion abilities well of existing methods. To address these limitations, we propose the R2C2-Coder to enhance and benchmark the real-world repository-level code completion abilities of code Large Language Models, where the R2C2-Coder includes a code prompt construction method R2C2-Enhance and a well-designed benchmark R2C2-Bench. Specifically, first, in R2C2-Enhance, we first construct the candidate retrieval pool and then assemble the completion prompt by retrieving from the retrieval pool for each completion cursor position. Second, based on R2C2 -Enhance, we can construct a more challenging and diverse R2C2-Bench with training, validation and test splits, where a context perturbation strategy is proposed to simulate the real-world repository-level code completion well. Extensive results on multiple benchmarks demonstrate the effectiveness of our R2C2-Coder.

Revealing hydrothermal evolution from the early oxide to late carbonate stages for quartz–wolframite vein-type deposits is essential for understanding the ore-forming process. In this study, we choose the Baishitouwa tungsten polymetallic deposit located in the southern Great Xing'an Range tungsten belt as a case study, and present detailed deposit geology and in situ fluid inclusion (FI) analyses including microthermometry, laser Raman spectra, and LA-ICP-MS microanalysis to address this issue. Four stages of hydrothermal activity were identified: (1) quartz–wolframite (I), (2) quartz–wolframite (II)–pyrite–chalcopyrite, (3) quartz–polymetallic sulphides, and (4) quartz–carbonate. Four types of FIs were recognized: CO2-rich, CO2-bearing, liquid-rich, and brine inclusions. Microthermometric data showed that the homogenization temperatures and salinities from the early to late stages are 380–460°C, 7.4–17.3, and 29.3–43.2 wt% NaCl equiv., 300–390°C and 7.1–17.0 wt% NaCl equiv., 220–320°C and 2.7–8.1 wt% NaCl equiv., and 150–250°C and 0.5–4.8 wt% NaCl equiv., respectively, suggesting a decreasing trend. Geochemically, all stage fluids contained high Rb and Mn concentrations, high Rb/Na, Cs/Na, Li/Na, K/Na, Rb/Sr, low K/Rb, and consistent Cs/Rb and Cs/(Na + K) ratios, indicating that the mineralizing fluids originated from a common source—an underlying, geochemically uniform, and highly fractionated granitic magma. Fluid immiscibility and cooling are the main mechanisms for wolframite precipitation, whereas greisenization is subordinate; the incursion of meteoric water into the hydrothermal system initiated at the sulphide stage, and fluid mixing is the dominant mechanism for sulphide precipitation.

Auxiliary clinical diagnosis has been researched to solve unevenly and insufficiently distributed clinical resources. However, auxiliary diagnosis is still dominated by human physicians, and how to make intelligent systems more involved in the diagnosis process is gradually becoming a concern. An interactive automated clinical diagnosis with a question-answering system and a question generation system can capture a patient's conditions from multiple perspectives with less physician involvement by asking different questions to drive and guide the diagnosis. This clinical diagnosis process requires diverse information to evaluate a patient from different perspectives to obtain an accurate diagnosis. Recently proposed medical question generation systems have not considered diversity. Thus, we propose a diversity learning-based visual question generation model using a multi-latent space to generate informative question sets from medical images. The proposed method generates various questions by embedding visual and language information in different latent spaces, whose diversity is trained by our newly proposed loss. We have also added control over the categories of generated questions, making the generated questions directional. Furthermore, we use a new metric named similarity to accurately evaluate the proposed model's performance. The experimental results on the Slake and VQA-RAD datasets demonstrate that the proposed method can generate questions with diverse information. Our model works with an answering model for interactive automated clinical diagnosis and generates datasets to replace the process of annotation that incurs huge labor costs.

Auxiliary clinical diagnosis has been researched to solve unevenly and insufficiently distributed clinical resources. However, auxiliary diagnosis is still dominated by human physicians, and how to make intelligent systems more involved in the diagnosis process is gradually becoming a concern. An interactive automated clinical diagnosis with a question-answering system and a question generation system can capture a patient's conditions from multiple perspectives with less physician involvement by asking different questions to drive and guide the diagnosis. This clinical diagnosis process requires diverse information to evaluate a patient from different perspectives to obtain an accurate diagnosis. Recently proposed medical question generation systems have not considered diversity. Thus, we propose a diversity learning-based visual question generation model using a multi-latent space to generate informative question sets from medical images. The proposed method generates various questions by embedding visual and language information in different latent spaces, whose diversity is trained by our newly proposed loss. We have also added control over the categories of generated questions, making the generated questions directional. Furthermore, we use a new metric named similarity to accurately evaluate the proposed model's performance. The experimental results on the Slake and VQA-RAD datasets demonstrate that the proposed method can generate questions with diverse information. Our model works with an answering model for interactive automated clinical diagnosis and generates datasets to replace the process of annotation that incurs huge labor costs.

BACKGROUND: The extracellular space between the cell wall and plasma membrane is a battlefield in plant-pathogen interactions. Within this space, the pathogen employs its secretome to attack the host in a variety of ways, including immunity manipulation. However, the role of the plant secretome is rarely studied for its role in disease resistance. RESULTS: Here, we examined the secretome of Verticillium wilt-resistant Gossypium hirsutum cultivar Zhongzhimian No.2 (ZZM2, encoding 95,327 predicted coding sequences) to determine its role in disease resistance against the wilt causal agent, Verticillium dahliae. Bioinformatics-driven analyses showed that the ZZM2 genome encodes 2085 secreted proteins and that these display disequilibrium in their distribution among the chromosomes. The cotton secretome displayed differences in the abundance of certain amino acid residues as compared to the remaining encoded proteins due to the localization of these putative proteins in the extracellular space. The secretome analysis revealed conservation for an allotetraploid genome, which nevertheless exhibited variation among orthologs and comparable unique genes between the two sub-genomes. Secretome annotation strongly suggested its involvement in extracellular stress responses (hydrolase activity, oxidoreductase activity, and extracellular region, etc.), thus contributing to resistance against the V. dahliae infection. Furthermore, the defense response genes (immunity marker NbHIN1, salicylic acid marker NbPR1, and jasmonic acid marker NbLOX4) were activated to varying degrees when Nicotina benthamiana leaves were agro-infiltrated with 28 randomly selected members, suggesting that the secretome plays an important role in the immunity response. Finally, gene silencing assays of 11 members from 13 selected candidates in ZZM2 displayed higher susceptibility to V. dahliae, suggesting that the secretome members confer the Verticillium wilt resistance in cotton. CONCLUSIONS: Our data dem

Auxiliary clinical diagnosis has been researched to solve unevenly and insufficiently distributed clinical resources. However, auxiliary diagnosis is still dominated by human physicians, and how to make intelligent systems more involved in the diagnosis process is gradually becoming a concern. An interactive automated clinical diagnosis with a question-answering system and a question generation system can capture a patient's conditions from multiple perspectives with less physician involvement by asking different questions to drive and guide the diagnosis. This clinical diagnosis process requires diverse information to evaluate a patient from different perspectives to obtain an accurate diagnosis. Recently proposed medical question generation systems have not considered diversity. Thus, we propose a diversity learning-based visual question generation model using a multi-latent space to generate informative question sets from medical images. The proposed method generates various questions by embedding visual and language information in different latent spaces, whose diversity is trained by our newly proposed loss. We have also added control over the categories of generated questions, making the generated questions directional. Furthermore, we use a new metric named similarity to accurately evaluate the proposed model's performance. The experimental results on the Slake and VQA-RAD datasets demonstrate that the proposed method can generate questions with diverse information. Our model works with an answering model for interactive automated clinical diagnosis and generates datasets to replace the process of annotation that incurs huge labor costs.

He Zhu,1,2,* Mujun Chang,1,3,* Qiulan Wang,1 Jing Chen,1 Dong Liu,1 Wenxi He1 1Department of Pharmacy, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, Hubei Province, People’s Republic of China; 2The Center for Biomedical Research, NHC Key Laboratory of Respiratory Diseases, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, Hubei Province, People’s Republic of China; 3Center for Translational Medicine, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, Hubei Province, People’s Republic of China*These authors contributed equally to this workCorrespondence: Dong Liu; Wenxi He, Department of Pharmacy, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, No. 1095 Jiefang Avenue, Wuhan, Hubei, People’s Republic of China, Email ld2069@outlook.com; wxhe@tjh.tjmu.edu.cnIntroduction: Pathogenic respiratory RNA viruses, including influenza A virus (IAV), respiratory syncytial virus (RSV), and SARS-CoV-2, are major causes of causes of acute respiratory infection globally. Plant-derived exosome-like nanoparticles containing miRNAs have shown substantial cross-kingdom regulatory effects on both viral and human transcripts. Houttuynia cordata (H. cordata), a traditional Chinese medicine frequently used to treat respiratory diseases. However, the role of H. cordata-derived exosome-like nanoparticles (HELNs) and the miRNA they encapsulated are unclear.Methods: HELNs were isolated from fresh underground roots (uHELNs) and above ground stems and leaves (aHELNs) using differential centrifugation. The HELNs were identified using transmission electron microscopy, nanoparticle tracking analysis, and zeta potential. Small RNA sequencing and RT-PCR were employed to determine the miRNA expression in uHELNs and aHELNs. All genomes were sourced from the NCBI database. Target prediction of vira

Xiaohui Wang,1,* Bo Wu,2,* Guogen Sun,3 Wenxi He,4 Jia Gao,1 Teng Huang,1 Jing Liu,1 Qing Zhou,1 Xiaoyu He,5 Shu Zhang,1 Zixiong Zhang,2 He Zhu1 1The Center for Biomedical Research, Department of Respiratory and Critical Care Medicine, NHC Key Laboratory of Respiratory Diseases, Tongji Hospital, Tongji Medical College, Huazhong University of Sciences and Technology, Wuhan, Hubei, People’s Republic of China; 2The Central Hospital of Enshi Tujia and Miao Autonomous Prefecture, Hubei Selenium and Human Health Institute, Enshi, Hubei, People’s Republic of China; 3Hubei Selenium and Human Health Institute, Enshi, Hubei, People’s Republic of China; 4Department of Pharmacy, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, Hubei Province, People’s Republic of China; 5Branch of National Clinical Research Center for Metabolic Diseases, Department of Endocrinology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, Hubei, People’s Republic of China*These authors contributed equally to this workCorrespondence: Zixiong Zhang, The Central Hospital of Enshi Tujia and Miao Autonomous Prefecture, Hubei Selenium and Human Health Institute, No. 158, Muyang Avenue, Enshi, Hubei, People’s Republic of China, Email 596099670@qq.com He Zhu, The Center for Biomedical Research, Department of Respiratory and Critical Care Medicine, NHC Key Laboratory of Respiratory Diseases, Tongji Hospital, Tongji Medical College, Huazhong University of Sciences and Technology, No. 1095 Jiefang Avenue, Wuhan, Hubei, People’s Republic of China, Email Zhuh@tjh.tjmu.edu.cnPurpose: Pancreatic adenocarcinoma (PAAD) presents an extremely high morbidity and mortality rate. Broccoli has excellent anti-cancer properties. However, the dosage and serious side effects still limit the application of broccoli and its derivatives for cancer therapy. Recently, extracellular vesicles (EVs) derived f

Recently, more and more people study online for the convenience of access to massive learning materials (e.g. test questions/notes), thus accurately understanding learning materials became a crucial issue, which is essential for many educational applications. Previous studies focus on using language models to represent the question data. However, test questions (TQ) are usually heterogeneous and multi-modal, e.g., some of them may only contain text, while others half contain images with information beyond their literal description. In this context, both supervised and unsupervised methods are difficult to learn a fused representation of questions. Meanwhile, this problem cannot be solved by conventional methods such as image caption, as the images may contain information complementary rather than duplicate to the text. In this paper, we first improve previous text-only representation with a two-stage unsupervised instance level contrastive based pre-training method (MCL: Mixture Unsupervised Contrastive Learning). Then, TQ-Net was proposed to fuse the content of images to the representation of heterogeneous data. Finally, supervised contrastive learning was conducted on relevance prediction-related downstream tasks, which helped the model to learn the representation of questions effectively. We conducted extensive experiments on question-based tasks on large-scale, real-world datasets, which demonstrated the effectiveness of TQ-Net and improve the precision of downstream applications (e.g. similar questions +2.02% and knowledge point prediction +7.20%). Our code will be available, and we will open-source a subset of our data to promote the development of relative studies., Comment: This paper has been accepted for the AAAI2023 AI4Edu Workshop

We present a novel multimodal interpretable VQA model that can answer the question more accurately and generate diverse explanations. Although researchers have proposed several methods that can generate human-readable and fine-grained natural language sentences to explain a model's decision, these methods have focused solely on the information in the image. Ideally, the model should refer to various information inside and outside the image to correctly generate explanations, just as we use background knowledge daily. The proposed method incorporates information from outside knowledge and multiple image captions to increase the diversity of information available to the model. The contribution of this paper is to construct an interpretable visual question answering model using multimodal inputs to improve the rationality of generated results. Experimental results show that our model can outperform state-of-the-art methods regarding answer accuracy and explanation rationality., Comment: Under review

The lifespan of lithium-ion batteries varies enormously from fundamental study to practical applications. This big difference has been typically ascribed to the high degree of uncertainty in unpredictable and complicated operation conditions in real-life applications. Here, we report that the pause of the charging-discharging process, which is frequently operated in practice but rarely studied in academics, is an important reason for the performance degradation of the NCM111 cathode. It is found that the pause during cycling could trigger a remarkable drop in capacity, giving rise to ∼30% more capacity decay compared with the continuously cycled sample. In situ synchrotron X-ray diffraction analysis reveals that the harmful H1-H2 phase transition, which typically appears in the initial cycle but disappears in subsequent cycles, is reactivated by the pausing process. The anisotropic lattice strains that occur during the H1-H2 transition result in mechanical fractures that terminate with an inert NiO-type rock-salt phase on the surface of particles. The present study indicates that the discontinuous usage of rechargeable batteries is also a key factor for cycle life, which might provide a distinct perspective on the performance decay in practical applications. © 2023 American Chemical Society.

Attention has become a common ingredient in deep learning architectures. It adds a dynamical selection of information on top of the static selection of information supported by weights. In the same way, we can imagine a higher-order informational filter built on top of attention: an Attention Schema (AS), namely, a descriptive and predictive model of attention. In cognitive neuroscience, Attention Schema Theory (AST) supports this idea of distinguishing attention from AS. A strong prediction of this theory is that an agent can use its own AS to also infer the states of other agents' attention and consequently enhance coordination with other agents. As such, multi-agent reinforcement learning would be an ideal setting to experimentally test the validity of AST. We explore different ways in which attention and AS interact with each other. Our preliminary results indicate that agents that implement the AS as a recurrent internal control achieve the best performance. In general, these exploratory experiments suggest that equipping artificial agents with a model of attention can enhance their social intelligence.

Hierarchical text classification (HTC) is a challenging subtask of multi-label classification as the labels form a complex hierarchical structure. Existing dual-encoder methods in HTC achieve weak performance gains with huge memory overheads and their structure encoders heavily rely on domain knowledge. Under such observation, we tend to investigate the feasibility of a memory-friendly model with strong generalization capability that could boost the performance of HTC without prior statistics or label semantics. In this paper, we propose Hierarchy-aware Tree Isomorphism Network (HiTIN) to enhance the text representations with only syntactic information of the label hierarchy. Specifically, we convert the label hierarchy into an unweighted tree structure, termed coding tree, with the guidance of structural entropy. Then we design a structure encoder to incorporate hierarchy-aware information in the coding tree into text representations. Besides the text encoder, HiTIN only contains a few multi-layer perceptions and linear transformations, which greatly saves memory. We conduct experiments on three commonly used datasets and the results demonstrate that HiTIN could achieve better test performance and less memory consumption than state-of-the-art (SOTA) methods., Comment: Accepted by ACL'23

In computer vision, contrastive learning is the most advanced unsupervised learning framework. Yet most previous methods simply apply fixed composition of data augmentations to improve data efficiency, which ignores the changes in their optimal settings over training. Thus, the pre-determined parameters of augmentation operations cannot always fit well with an evolving network during the whole training period, which degrades the quality of the learned representations. In this work, we propose AdDA, which implements a closed-loop feedback structure to a generic contrastive learning network. AdDA works by allowing the network to adaptively adjust the augmentation compositions according to the real-time feedback. This online adjustment helps maintain the dynamic optimal composition and enables the network to acquire more generalizable representations with minimal computational overhead. AdDA achieves competitive results under the common linear protocol on ImageNet-100 classification (+1.11% on MoCo v2)., Comment: Accepted by ICASSP 2023 (Oral)

New physical insights and a robust approach are provided to develop super-MIEC anodes for high-rate batteries, which would bridge two materials families—nanoporous framework materials and conductive oxides.

This paper addresses the challenge of 3D instance segmentation by simultaneously leveraging 3D geometric and multi-view image information. Many previous works have applied deep learning techniques to 3D point clouds for instance segmentation. However, these methods often failed to generalize to various types of scenes due to the scarcity and low-diversity of labeled 3D point cloud data. Some recent works have attempted to lift 2D instance segmentations to 3D within a bottom-up framework. The inconsistency in 2D instance segmentations among views can substantially degrade the performance of 3D segmentation. In this work, we introduce a novel 3D-to-2D query framework to effectively exploit 2D segmentation models for 3D instance segmentation. Specifically, we pre-segment the scene into several superpoints in 3D, formulating the task into a graph cut problem. The superpoint graph is constructed based on 2D segmentation models, where node features are obtained from multi-view image features and edge weights are computed based on multi-view segmentation results, enabling the better generalization ability. To process the graph, we train a graph neural network using pseudo 3D labels from 2D segmentation models. Experimental results on the ScanNet, ScanNet++ and KITTI-360 datasets demonstrate that our method achieves robust segmentation performance and can generalize across different types of scenes. Our project page is available at https://zju3dv.github.io/sam_graph., Comment: Project page: https://zju3dv.github.io/sam_graph

The SoccerNet 2022 challenges were the second annual video understanding challenges organized by the SoccerNet team. In 2022, the challenges were composed of 6 vision-based tasks: (1) action spotting, focusing on retrieving action timestamps in long untrimmed videos, (2) replay grounding, focusing on retrieving the live moment of an action shown in a replay, (3) pitch localization, focusing on detecting line and goal part elements, (4) camera calibration, dedicated to retrieving the intrinsic and extrinsic camera parameters, (5) player re-identification, focusing on retrieving the same players across multiple views, and (6) multiple object tracking, focusing on tracking players and the ball through unedited video streams. Compared to last year's challenges, tasks (1-2) had their evaluation metrics redefined to consider tighter temporal accuracies, and tasks (3-6) were novel, including their underlying data and annotations. More information on the tasks, challenges and leaderboards are available on https://www.soccer-net.org. Baselines and development kits are available on https://github.com/SoccerNet.

Machine vision systems (MVSs) are an important component of intelligent systems, such as autonomous vehicles and robots. However, with the continuous increase in data and new application scenarios, new requirements are put forward for the next generation of MVS. There is an urgent need to find new material systems to complement the existing semiconductor technology based on thin-film materials, and new architectures must be explored to improve efficiency. Because of their unique physical properties, two-dimensional (2D) materials have received extensive attention for use in MVSs, especially in biomimetic ones: the human visual system, which can process complex visual information with low power consumption, provides a model for next-generation MVSs. This review paper summarizes the progress and challenges of applying 2D material photodetectors in sense-memory-computational integration and biomimetic image sensors for machine vision. (Figure presented.). © 2021 The Authors. InfoMat published by UESTC and John Wiley & Sons Australia, Ltd.

BackgroundDuring the disease cycle, plant pathogenic fungi exhibit a morphological transition between hyphal growth (the phase of active infection) and the production of long-term survival structures that remain dormant during "overwintering." Verticillium dahliae is a major plant pathogen that produces heavily melanized microsclerotia (MS) that survive in the soil for 14 or more years. These MS are multicellular structures produced during the necrotrophic phase of the disease cycle. Polyketide synthases (PKSs) are responsible for catalyzing production of many secondary metabolites including melanin. While MS contribute to long-term survival, hyphal growth is key for infection and virulence, but the signaling mechanisms by which the pathogen maintains hyphal growth are unclear.ResultsWe analyzed the VdPKSs that contain at least one conserved domain potentially involved in secondary metabolism (SM), and screened the effect of VdPKS deletions in the virulent strain AT13. Among the five VdPKSs whose deletion affected virulence on cotton, we found that VdPKS9 acted epistatically to the VdPKS1-associated melanin pathway to promote hyphal growth. The decreased hyphal growth in VdPKS9 mutants was accompanied by the up-regulation of melanin biosynthesis and MS formation. Overexpression of VdPKS9 transformed melanized hyphal-type (MH-type) into the albinistic hyaline hyphal-type (AH-type), and VdPKS9 was upregulated in the AH-type population, which also exhibited higher virulence than the MH-type.ConclusionsWe show that VdPKS9 is a powerful negative regulator of both melanin biosynthesis and MS formation in V. dahliae. These findings provide insight into the mechanism of how plant pathogens promote their virulence by the maintenance of vegetative hyphal growth during infection and colonization of plant hosts, and may provide novel targets for the control of melanin-producing filamentous fungi.

The graduating class of the MFA Fashion Design program presented their final pieces in a runway show on Wednesday, June 8 at 3 pm in the Katie Murphy Amphitheatre.

2022 Fashion Design MFA graduate thesis collection created by He (Irene) Zhu. Presented on June 8, 2022 at 'Unbound: Fashion Design MFA Showcase.', Department: Fashion Design., Chinese designer He Zhu's philosophy is that "Fashion is a form of fine art." Her collection "cloth: square: white" is a way to experiment with clothing as an extension of the spiritual body. The collection projects the ancient East Asian view of the body into contemporary fashion through light & shadow and geometry. The designs feature white squares in different variations and combinations that change the silhouette on the body, with an aim to create a sustainable lifestyle - a unity between man and universe., He Zhu pursues fashion as a fine art form that also incorporates modern ethos. Once trained as a scientist, Zhu is dedicated to creating a solid theoretical foundation for her design with a precise touch of details. Zhu inherited her grandma's aesthetic taste: the kind of elegance and subtlety that comes from the Chinese scholar-gentry in the early 19th century. With an aim to Introduce a balance of Apollo and Dionysian spirits into her designs, she explores the extension of the body with a sustainable approach – folding, zero waste patterns, collaging – reimagining fashion from linear to spatial, sensorial, and temporal dimensions.

Verticillium dahliae is a destructive soil-borne pathogen of many economically important dicots. The genetics of pathogenesis in V. dahliae has been extensively studied. Spt-Ada-Gcn5 acetyltransferase complex (SAGA) is an ATP-independent multifunctional chromatin remodeling complex that contributes to diverse transcriptional regulatory functions. As members of the core module in the SAGA complex in Saccharomyces cerevisiae, Ada1, together with Spt7 and Spt20, play an important role in maintaining the integrity of the complex. In this study, we identified homologs of the SAGA complex in V. dahliae and found that deletion of the Ada1 subunit (VdAda1) causes severe defects in the formation of conidia and microsclerotia, and in melanin biosynthesis and virulence. The effect of VdAda1 on histone acetylation in V. dahliae was confirmed by western blot analysis. The deletion of VdAda1 resulted in genome-wide alteration of the V. dahliae transcriptome, including genes encoding transcription factors and secreted proteins, suggesting its prominent role in the regulation of transcription and virulence. Overall, we demonstrated that VdAda1, a member of the SAGA complex, modulates multiple physiological processes by regulating global gene expression that impinge on virulence and survival in V. dahliae.

BackgroundDuring the disease cycle, plant pathogenic fungi exhibit a morphological transition between hyphal growth (the phase of active infection) and the production of long-term survival structures that remain dormant during "overwintering." Verticillium dahliae is a major plant pathogen that produces heavily melanized microsclerotia (MS) that survive in the soil for 14 or more years. These MS are multicellular structures produced during the necrotrophic phase of the disease cycle. Polyketide synthases (PKSs) are responsible for catalyzing production of many secondary metabolites including melanin. While MS contribute to long-term survival, hyphal growth is key for infection and virulence, but the signaling mechanisms by which the pathogen maintains hyphal growth are unclear.ResultsWe analyzed the VdPKSs that contain at least one conserved domain potentially involved in secondary metabolism (SM), and screened the effect of VdPKS deletions in the virulent strain AT13. Among the five VdPKSs whose deletion affected virulence on cotton, we found that VdPKS9 acted epistatically to the VdPKS1-associated melanin pathway to promote hyphal growth. The decreased hyphal growth in VdPKS9 mutants was accompanied by the up-regulation of melanin biosynthesis and MS formation. Overexpression of VdPKS9 transformed melanized hyphal-type (MH-type) into the albinistic hyaline hyphal-type (AH-type), and VdPKS9 was upregulated in the AH-type population, which also exhibited higher virulence than the MH-type.ConclusionsWe show that VdPKS9 is a powerful negative regulator of both melanin biosynthesis and MS formation in V. dahliae. These findings provide insight into the mechanism of how plant pathogens promote their virulence by the maintenance of vegetative hyphal growth during infection and colonization of plant hosts, and may provide novel targets for the control of melanin-producing filamentous fungi.

The SoccerNet 2022 challenges were the second annual video understanding challenges organized by the SoccerNet team. In 2022, the challenges were composed of 6 vision-based tasks: (1) action spotting, focusing on retrieving action timestamps in long untrimmed videos, (2) replay grounding, focusing on retrieving the live moment of an action shown in a replay, (3) pitch localization, focusing on detecting line and goal part elements, (4) camera calibration, dedicated to retrieving the intrinsic and extrinsic camera parameters, (5) player re-identification, focusing on retrieving the same players across multiple views, and (6) multiple object tracking, focusing on tracking players and the ball through unedited video streams. Compared to last year's challenges, tasks (1-2) had their evaluation metrics redefined to consider tighter temporal accuracies, and tasks (3-6) were novel, including their underlying data and annotations. More information on the tasks, challenges and leaderboards are available on https://www.soccer-net.org. Baselines and development kits are available on https://github.com/SoccerNet., Comment: Accepted at ACM MMSports 2022

Human intelligence is able to first learn some basic skills for solving basic problems and then assemble such basic skills into complex skills for solving complex or new problems. For example, the basic skills "dig hole," "put tree," "backfill" and "watering" compose a complex skill "plant a tree". Besides, some basic skills can be reused for solving other problems. For example, the basic skill "dig hole" not only can be used for planting a tree, but also can be used for mining treasures, building a drain, or landfilling. The ability to learn basic skills and reuse them for various tasks is very important for humans because it helps to avoid learning too many skills for solving each individual task, and makes it possible to solve a compositional number of tasks by learning just a few number of basic skills, which saves a considerable amount of memory and computation in the human brain. We believe that machine intelligence should also capture the ability of learning basic skills and reusing them by composing into complex skills. In computer science language, each basic skill is a "module", which is a reusable network of a concrete meaning and performs a specific basic operation. The modules are assembled into a bigger "model" for doing a more complex task. The assembling procedure is adaptive to the input or task, i.e., for a given task, the modules should be assembled into the best model for solving the task. As a result, different inputs or tasks could have different assembled models, which enables Auto-Assembling AI (AAAI). In this work, we propose Modularized Adaptive Neural Architecture Search (MANAS) to demonstrate the above idea. Experiments on different datasets show that the adaptive architecture assembled by MANAS outperforms static global architectures. Further experiments and empirical analysis provide insights to the effectiveness of MANAS., Comment: In ACM CIKM 2022

Lorises are a group of globally threatened strepsirrhine primates that exhibit many unusual physiological and behavioral features, including a low metabolic rate, slow movement, and hibernation. Here, we assembled a chromosome-level genome sequence of the pygmy loris (Xanthonycticebus pygmaeus) and resequenced whole genomes from 50 pygmy lorises and 6 Bengal slow lorises (Nycticebus bengalensis). We found that many gene families involved in detoxification have been specifically expanded in the pygmy loris, including the GSTA gene family, with many newly derived copies functioning specifically in the liver. We detected many genes displaying evolutionary convergence between pygmy loris and koala, including PITRM1. Significant decreases in PITRM1 enzymatic activity in these two species may have contributed to their characteristic low rate of metabolism. We also detected many evolutionarily convergent genes and positively selected genes in the pygmy loris that are involved in muscle development. Functional assays demonstrated the decreased ability of one positively selected gene, MYOF, to up-regulate the fast-type muscle fiber, consistent with the lower proportion of fast-twitch muscle fibers in the pygmy loris. The protein product of another positively selected gene in the pygmy loris, PER2, exhibited weaker binding to the key circadian core protein CRY, a finding that may be related to this species’ unusual circadian rhythm. Finally, population genomics analysis revealed that these two extant loris species, which coexist in the same habitat, have exhibited an inverse relationship in terms of their demography over the past 1 million years, implying strong interspecies competition after speciation.

Lorises are a group of globally threatened strepsirrhine primates that exhibit many unusual physiological and behavioral features, including a low metabolic rate, slow movement, and hibernation. Here, we assembled a chromosome-level genome sequence of the pygmy loris (Xanthonycticebus pygmaeus) and resequenced whole genomes from 50 pygmy lorises and 6 Bengal slow lorises (Nycticebus bengalensis). We found that many gene families involved in detoxification have been specifically expanded in the pygmy loris, including the GSTA gene family, with many newly derived copies functioning specifically in the liver. We detected many genes displaying evolutionary convergence between pygmy loris and koala, including PITRM1. Significant decreases in PITRM1 enzymatic activity in these two species may have contributed to their characteristic low rate of metabolism. We also detected many evolutionarily convergent genes and positively selected genes in the pygmy loris that are involved in muscle development. Functional assays demonstrated the decreased ability of one positively selected gene, MYOF, to up-regulate the fast-type muscle fiber, consistent with the lower proportion of fast-twitch muscle fibers in the pygmy loris. The protein product of another positively selected gene in the pygmy loris, PER2, exhibited weaker binding to the key circadian core protein CRY, a finding that may be related to this species’ unusual circadian rhythm. Finally, population genomics analysis revealed that these two extant loris species, which coexist in the same habitat, have exhibited an inverse relationship in terms of their demography over the past 1 million years, implying strong interspecies competition after speciation.

2022 Fashion Design MFA graduate thesis collection created by He (Irene) Zhu. Presented on June 8, 2022 at 'Unbound: Fashion Design MFA Showcase.', Department: Fashion Design., Chinese designer He Zhu's philosophy is that "Fashion is a form of fine art." Her collection "cloth: square: white" is a way to experiment with clothing as an extension of the spiritual body. The collection projects the ancient East Asian view of the body into contemporary fashion through light & shadow and geometry. The designs feature white squares in different variations and combinations that change the silhouette on the body, with an aim to create a sustainable lifestyle - a unity between man and universe., He Zhu pursues fashion as a fine art form that also incorporates modern ethos. Once trained as a scientist, Zhu is dedicated to creating a solid theoretical foundation for her design with a precise touch of details. Zhu inherited her grandma's aesthetic taste: the kind of elegance and subtlety that comes from the Chinese scholar-gentry in the early 19th century. With an aim to Introduce a balance of Apollo and Dionysian spirits into her designs, she explores the extension of the body with a sustainable approach – folding, zero waste patterns, collaging – reimagining fashion from linear to spatial, sensorial, and temporal dimensions.

The graduating class of the MFA Fashion Design program presented their final pieces in a runway show on Wednesday, June 8 at 3 pm in the Katie Murphy Amphitheatre.

Proton conduction underlies many important electrochemical technologies. A family of new proton electrolytes is reported: acid-in-clay electrolyte (AiCE) prepared by integrating fast proton carriers in a natural phyllosilicate clay network, which can be made into thin-film (tens of micrometers) fluid-impervious membranes. The chosen example systems (sepiolite-phosphoric acid) rank top among the solid proton conductors in terms of proton conductivities (15 mS cm-1 at 25 °C, 0.023 mS cm-1 at -82 °C), electrochemical stability window (3.35 V), and reduced chemical reactivity. A proton battery is assembled using AiCE as the solid electrolyte membrane. Benefitting from the wider electrochemical stability window, reduced corrosivity, and excellent ionic selectivity of AiCE, the two main problems (gassing and cyclability) of proton batteries are successfully solved. This work draws attention to the element cross-over problem in proton batteries and the generic "acid-in-clay" solid electrolyte approach with superfast proton transport, outstanding selectivity, and improved stability for room- to cryogenic-temperature protonic applications.

This paper describes the design and flight test of the control system of a novel octocopter with new topology to improve aerodynamic performance and efficiency. The article analyzes the advantages of the new configuration and analyzes the feasibility of the control method which can ensure fault-tolerant control when one rotor of the aircraft stops in theory. The feasibility verification is carried out through a prototype flight test. The power and hardware platform of the octocopter UAV was set up according to the requirements of structural and control system design. Flight tests were carried out multiple times, and the UAV’s redundant actuation was able to successfully stabilize the vehicle, even after a single rotor stopped functioning. Under normal flight conditions, smooth flight and effective control could be ensured for the UAV, while under the condition of a single rotor having stopped functioning, the control method was able to effectively utilize the other seven rotors to provide a proper lift force and to control the aircraft to perform basic motions such as pitch, roll, and yaw.

Neural network policies trained using Deep Reinforcement Learning (DRL) are well-known to be susceptible to adversarial attacks. In this paper, we consider attacks manifesting as perturbations in the observation space managed by the external environment. These attacks have been shown to downgrade policy performance significantly. We focus our attention on well-trained deterministic and stochastic neural network policies in the context of continuous control benchmarks subject to four well-studied observation space adversarial attacks. To defend against these attacks, we propose a novel defense strategy using a detect-and-denoise schema. Unlike previous adversarial training approaches that sample data in adversarial scenarios, our solution does not require sampling data in an environment under attack, thereby greatly reducing risk during training. Detailed experimental results show that our technique is comparable with state-of-the-art adversarial training approaches.

Machine vision systems (MVSs) are an important component of intelligent systems, such as autonomous vehicles and robots. However, with the continuous increase in data and new application scenarios, new requirements are put forward for the next generation of MVS. There is an urgent need to find new material systems to complement the existing semiconductor technology based on thin-film materials, and new architectures must be explored to improve efficiency. Because of their unique physical properties, two-dimensional (2D) materials have received extensive attention for use in MVSs, especially in biomimetic ones: the human visual system, which can process complex visual information with low power consumption, provides a model for next-generation MVSs. This review paper summarizes the progress and challenges of applying 2D material photodetectors in sense-memory-computational integration and biomimetic image sensors for machine vision. (Figure presented.). © 2021 The Authors. InfoMat published by UESTC and John Wiley & Sons Australia, Ltd.

O-GalNAc glycans (or mucin O-glycans) play pivotal roles in diverse biological and pathological processes, including tumor growth and progression. Structurally defined O-GalNAc glycans are essential for functional studies but synthetic challenges and their inherent structural diversity and complexity have limited access to these compounds. Herein, we report an efficient and robust chemoenzymatic modular assembly (CEMA) strategy to construct structurally diverse O-GalNAc glycans. The key to this strategy is the convergent assembly of O-GalNAc cores 1-4 and 6 from three chemical building blocks, followed by enzymatic diversification of the cores by 13 well-tailored enzyme modules. A total of 83 O-GalNAc glycans presenting various natural glycan epitopes are obtained and used to generate a unique synthetic mucin O-glycan microarray. Binding specificities of glycan-binding proteins (GBPs) including plant lectins and selected anti-glycan antibodies towards these O-GalNAc glycans are revealed by this microarray, promoting their applicability in functional O-glycomics. Serum samples from colorectal cancer patients and healthy controls are assayed using the array reveal higher bindings towards less common cores 3, 4, and 6 than abundant cores 1 and 2, providing insights into O-GalNAc glycan structure-activity relationships.

© 2021 The Nonferrous Metals Society of China MnFeCoCuNix high-entropy alloys (HEAs) with different Ni contents were fabricated by vacuum induction melting. XRD and SEM−EDS were used to analyze the phase constitution and structure, and the tensile properties of the samples were determined using a universal tensile tester. The results show that the HEAs consist of a dual-phase structure, in which FCC1 phase is rich in Fe and Co, while the FCC2 phase has high contents of Cu and Mn. As Ni content increases, the segregation of Cu decreases, accompanied by the decrease of FCC2 phase. Moreover, the tensile strength of the HEAs increases first and then decreases, and the elongation increases slightly. This is attributed to the combined effect of interface strengthening and solid solution strengthening. The in-situ stretched MnFeCoCuNi0.5 alloy shows obvious neck shrinkage during the tensile fracture process. In the initial deformation stage, the slip lines show different morphologies in the dual-phase structure. However, in the later stage, the surface slip lines become longer and denser due to the redistribution of atoms and the re-separation of the dissolved phase.

O-GalNAc glycans (or mucin O-glycans) play pivotal roles in diverse biological and pathological processes, including tumor growth and progression. Structurally defined O-GalNAc glycans are essential for functional studies but synthetic challenges and their inherent structural diversity and complexity have limited access to these compounds. Herein, we report an efficient and robust chemoenzymatic modular assembly (CEMA) strategy to construct structurally diverse O-GalNAc glycans. The key to this strategy is the convergent assembly of O-GalNAc cores 1-4 and 6 from three chemical building blocks, followed by enzymatic diversification of the cores by 13 well-tailored enzyme modules. A total of 83 O-GalNAc glycans presenting various natural glycan epitopes are obtained and used to generate a unique synthetic mucin O-glycan microarray. Binding specificities of glycan-binding proteins (GBPs) including plant lectins and selected anti-glycan antibodies towards these O-GalNAc glycans are revealed by this microarray, promoting their applicability in functional O-glycomics. Serum samples from colorectal cancer patients and healthy controls are assayed using the array reveal higher bindings towards less common cores 3, 4, and 6 than abundant cores 1 and 2, providing insights into O-GalNAc glycan structure-activity relationships.

© 2021 The Nonferrous Metals Society of China MnFeCoCuNix high-entropy alloys (HEAs) with different Ni contents were fabricated by vacuum induction melting. XRD and SEM−EDS were used to analyze the phase constitution and structure, and the tensile properties of the samples were determined using a universal tensile tester. The results show that the HEAs consist of a dual-phase structure, in which FCC1 phase is rich in Fe and Co, while the FCC2 phase has high contents of Cu and Mn. As Ni content increases, the segregation of Cu decreases, accompanied by the decrease of FCC2 phase. Moreover, the tensile strength of the HEAs increases first and then decreases, and the elongation increases slightly. This is attributed to the combined effect of interface strengthening and solid solution strengthening. The in-situ stretched MnFeCoCuNi0.5 alloy shows obvious neck shrinkage during the tensile fracture process. In the initial deformation stage, the slip lines show different morphologies in the dual-phase structure. However, in the later stage, the surface slip lines become longer and denser due to the redistribution of atoms and the re-separation of the dissolved phase.

Graphs can represent relational information among entities and graph structures are widely used in many intelligent tasks such as search, recommendation, and question answering. However, most of the graph-structured data in practice suffers from incompleteness, and thus link prediction becomes an important research problem. Though many models are proposed for link prediction, the following two problems are still less explored: (1) Most methods model each link independently without making use of the rich information from relevant links, and (2) existing models are mostly designed based on associative learning and do not take reasoning into consideration. With these concerns, in this paper, we propose Graph Collaborative Reasoning (GCR), which can use the neighbor link information for relational reasoning on graphs from logical reasoning perspectives. We provide a simple approach to translate a graph structure into logical expressions, so that the link prediction task can be converted into a neural logic reasoning problem. We apply logical constrained neural modules to build the network architecture according to the logical expression and use back propagation to efficiently learn the model parameters, which bridges differentiable learning and symbolic reasoning in a unified architecture. To show the effectiveness of our work, we conduct experiments on graph-related tasks such as link prediction and recommendation based on commonly used benchmark datasets, and our graph collaborative reasoning approach achieves state-of-the-art performance., Comment: 10 pages. Accepted to ACM WSDM 2022

Counterfactual examples (CFs) are one of the most popular methods for attaching post-hoc explanations to machine learning (ML) models. However, existing CF generation methods either exploit the internals of specific models or depend on each sample's neighborhood, thus they are hard to generalize for complex models and inefficient for large datasets. This work aims to overcome these limitations and introduces ReLAX, a model-agnostic algorithm to generate optimal counterfactual explanations. Specifically, we formulate the problem of crafting CFs as a sequential decision-making task and then find the optimal CFs via deep reinforcement learning (DRL) with discrete-continuous hybrid action space. Extensive experiments conducted on several tabular datasets have shown that ReLAX outperforms existing CF generation baselines, as it produces sparser counterfactuals, is more scalable to complex target models to explain, and generalizes to both classification and regression tasks. Finally, to demonstrate the usefulness of our method in a real-world use case, we leverage CFs generated by ReLAX to suggest actions that a country should take to reduce the risk of mortality due to COVID-19. Interestingly enough, the actions recommended by our method correspond to the strategies that many countries have actually implemented to counter the COVID-19 pandemic.

Text classification is a primary task in natural language processing (NLP). Recently, graph neural networks (GNNs) have developed rapidly and been applied to text classification tasks. As a special kind of graph data, the tree has a simpler data structure and can provide rich hierarchical information for text classification. Inspired by the structural entropy, we construct the coding tree of the graph by minimizing the structural entropy and propose HINT, which aims to make full use of the hierarchical information contained in the text for the task of text classification. Specifically, we first establish a dependency parsing graph for each text. Then we designed a structural entropy minimization algorithm to decode the key information in the graph and convert each graph to its corresponding coding tree. Based on the hierarchical structure of the coding tree, the representation of the entire graph is obtained by updating the representation of non-leaf nodes in the coding tree layer by layer. Finally, we present the effectiveness of hierarchical information in text classification. Experimental results show that HINT outperforms the state-of-the-art methods on popular benchmarks while having a simple structure and few parameters., Comment: COLING 2022(Oral)

Wu F, Sui Y, Wang Y, Xu T, Fan L, Zhu H. Onco Targets Ther. 2020;13:685–699. The Editor and Publisher of OncoTargets and Therapy wish to retract the published article. Concerns were raised regarding the integrity of the tumor data presented in Figure 7B and the alleged duplication of images in Figure 2C with similar images from unrelated studies, specifically: Figure 2C panel C-33A, si-NC appears to have been duplicated with a similar image in Figure 2C from Gao et al, 2020 (https://doi.org/10.1080/15384101.2020.1749466). The authors responded to our queries but were unable to provide an explanation for the alleged image duplication and could not provide the original raw data for their study and have therefore agreed to retract the article. Our decision-making was informed by our policy on publishing ethics and integrity and the COPE guidelines on retraction. The retracted article will remain online to maintain the scholarly record, but it will be digitally watermarked on each page as “Retracted”. This retraction relates to this paper

Most classification models can be considered as the process of matching templates. However, when intra-class uncertainty/variability is not considered, especially for datasets containing unbalanced classes, this may lead to classification errors. To address this issue, we propose a loss function with intra-class uncertainty following Gaussian distribution. Specifically, in our framework, the features extracted by deep networks of each class are characterized by independent Gaussian distribution. The parameters of distribution are learned with a likelihood regularization along with other network parameters. The means of the Gaussian play a similar role as the center anchor in existing methods, and the variance describes the uncertainty of different classes. In addition, similar to the inter-class margin in traditional loss functions, we introduce a margin to intra-class uncertainty to make each cluster more compact and reduce the imbalance of feature distribution from different categories. Based on MNIST, CIFAR, ImageNet, and Long-tailed CIFAR analyses, the proposed approach shows improved classification performance, through learning a better class representation., Comment: Accepted to ICME 2021

Aolin Ren,1,* Na Zhang,1,* He Zhu,1 Kang Zhou,1 Yuan Cao,1 Jindong Liu1,2 1Department of Anesthesiology, Affiliated Hospital of Xuzhou Medical University, Xuzhou, Jiangsu, People’s Republic of China; 2Jiangsu Province Key Laboratory of Anesthesiology, Xuzhou Medical University, Xuzhou, Jiangsu, People’s Republic of China*These authors contributed equally to this workCorrespondence: Jindong LiuDepartment of Anesthesiology, Affiliated Hospital of Xuzhou Medical University, Huaihai West Road #99, Xuzhou, Jiangsu, 221000, People’s Republic of ChinaEmail liujindong1818@163.com View the original paper by Dr Ren and colleagues This is in response to the Letter to the Editor