Search

Marine seismic interference noise occurs when energy from nearby marine seismic source vessels is recorded during a seismic survey. Such noise tends to be well preserved over large distances and cause coherent artifacts in the recorded data. Over the years, the industry has developed various denoising techniques for seismic interference removal, but although well performing they are still time-consuming in use. Machine-learning based processing represents an alternative approach, which may significantly improve the computational efficiency. In case of conventional images, autoencoders are frequently employed for denoising purposes. However, due to the special characteristics of seismic data as well as the noise, autoencoders failed in the case of marine seismic interference noise. We therefore propose the use of a customized U-Net design with element-wise summation as part of the skip-connection blocks to handle the vanishing gradient problem and to ensure information fusion between high- and low-level features. To secure a realistic study, only seismic field data were employed, including 25000 training examples. The customized U-Net was found to perform well leaving only minor residuals, except for the case when seismic interference noise comes from the side. We further demonstrate that such noise can be treated by slightly increasing the depth of our network. Although our customized U-Net does not outperform a standard commercial algorithm in quality, it can (after proper training) read and process one single shot gather in approximately 0.02s. This is significantly faster than any existing industry denoising algorithm. In addition, the proposed network processes shot gathers in a sequential order, which is an advantage compared with industry algorithms that typically require a multi-shot input to break the coherency of the noise.

Clinical diagnosis is critical in medical practice, typically requiring a continuous and evolving process that includes primary diagnosis, differential diagnosis, and final diagnosis. However, most existing clinical diagnostic tasks are single-step processes, which does not align with the complex multi-step diagnostic procedures found in real-world clinical settings. In this paper, we propose a multi-step diagnostic task and annotate a clinical diagnostic dataset (MSDiagnosis). This dataset includes primary diagnosis, differential diagnosis, and final diagnosis questions. Additionally, we propose a novel and effective framework. This framework combines forward inference, backward inference, reflection, and refinement, enabling the LLM to self-evaluate and adjust its diagnostic results. To assess the effectiveness of our proposed method, we design and conduct extensive experiments. The experimental results demonstrate the effectiveness of the proposed method. We also provide a comprehensive experimental analysis and suggest future research directions for this task.

Few-shot action recognition is an emerging field in computer vision, primarily focused on meta-learning within the same domain. However, challenges arise in real-world scenario deployment, as gathering extensive labeled data within a specific domain is laborious and time-intensive. Thus, attention shifts towards cross-domain few-shot action recognition, requiring the model to generalize across domains with significant deviations. Therefore, we propose a novel approach, ``Distillation from Mixed-Source Domain", tailored to address this conundrum. Our method strategically integrates insights from both labeled data of the source domain and unlabeled data of the target domain during the training. The ResNet18 is used as the backbone to extract spatial features from the source and target domains. We design two branches for meta-training: the original-source and the mixed-source branches. In the first branch, a Domain Temporal Encoder is employed to capture temporal features for both the source and target domains. Additionally, a Domain Temporal Decoder is employed to reconstruct all extracted features. In the other branch, a Domain Mixed Encoder is used to handle labeled source domain data and unlabeled target domain data, generating mixed-source domain features. We incorporate a pre-training stage before meta-training, featuring a network architecture similar to that of the first branch. Lastly, we introduce a dual distillation mechanism to refine the classification probabilities of source domain features, aligning them with those of mixed-source domain features. This iterative process enriches the insights of the original-source branch with knowledge from the mixed-source branch, thereby enhancing the model's generalization capabilities. Our code is available at URL: \url{https://xxxx/xxxx/xxxx.git}

The determination of the speed of sound in quark-gluon plasma is a crucial aspect of understanding the properties of strongly interacting matter created in relativistic heavy-ion collisions. In this study, we investigate the impact of initial-state fluctuations on the extraction of the speed of sound in a quark-gluon plasma in the ultra-central collisions. By employing the TRENTo model for simulating initial conditions, we demonstrate that these fluctuations lead to sizable volume effect, which in turn corrects the measured values of the sound velocity. With respect to realistic conditions of high-energy heavy-ion experiments, we provide quantitative estimate of these corrections.

Large language models (LLMs) have revolutionized Natural Language Processing (NLP) by minimizing the need for complex feature engineering. However, the application of LLMs in specialized domains like biopharmaceuticals and chemistry remains largely unexplored. These fields are characterized by intricate terminologies, specialized knowledge, and a high demand for precision areas where general purpose LLMs often fall short. In this study, we introduce PharmaGPT, a suite of domain specilized LLMs with 13 billion and 70 billion parameters, specifically trained on a comprehensive corpus tailored to the Bio-Pharmaceutical and Chemical domains. Our evaluation shows that PharmaGPT surpasses existing general models on specific-domain benchmarks such as NAPLEX, demonstrating its exceptional capability in domain-specific tasks. Remarkably, this performance is achieved with a model that has only a fraction, sometimes just one-tenth-of the parameters of general-purpose large models. This advancement establishes a new benchmark for LLMs in the bio-pharmaceutical and chemical fields, addressing the existing gap in specialized language modeling. It also suggests a promising path for enhanced research and development, paving the way for more precise and effective NLP applications in these areas.

In recent years, large language models(LLMs) have attracted significant attention due to their exceptional performance across a multitude of natural language process tasks, and have been widely applied in various fields. However, the application of large language models in the Intellectual Property (IP) domain is challenging due to the strong need for specialized knowledge, privacy protection, processing of extremely long text in this field. In this technical report, we present for the first time a low-cost, standardized procedure for training IP-oriented LLMs, meeting the unique requirements of the IP domain. Using this standard process, we have trained the PatentGPT series models based on open-source pretrained models. By evaluating them on the open-source IP-oriented benchmark MOZIP, our domain-specific LLMs outperforms GPT-4, indicating the effectiveness of the proposed training procedure and the expertise of the PatentGPT models in the IP domain. Remarkably, our model surpassed GPT-4 on the 2019 China Patent Agent Qualification Examination, scoring 65 and matching human expert levels. Additionally, the PatentGPT model, which utilizes the SMoE architecture, achieves performance comparable to that of GPT-4 in the IP domain and demonstrates a better cost-performance ratio on long-text tasks, potentially serving as an alternative to GPT-4 within the IP domain., Comment: 19 pages, 9 figures

In recommender systems, multi-behavior methods have demonstrated their effectiveness in mitigating issues like data sparsity, a common challenge in traditional single-behavior recommendation approaches. These methods typically infer user preferences from various auxiliary behaviors and apply them to the target behavior for recommendations. However, this direct transfer can introduce noise to the target behavior in recommendation, due to variations in user attention across different behaviors. To address this issue, this paper introduces a novel approach, Behavior-Contextualized Item Preference Modeling (BCIPM), for multi-behavior recommendation. Our proposed Behavior-Contextualized Item Preference Network discerns and learns users' specific item preferences within each behavior. It then considers only those preferences relevant to the target behavior for final recommendations, significantly reducing noise from auxiliary behaviors. These auxiliary behaviors are utilized solely for training the network parameters, thereby refining the learning process without compromising the accuracy of the target behavior recommendations. To further enhance the effectiveness of BCIPM, we adopt a strategy of pre-training the initial embeddings. This step is crucial for enriching the item-aware preferences, particularly in scenarios where data related to the target behavior is sparse. Comprehensive experiments conducted on four real-world datasets demonstrate BCIPM's superior performance compared to several leading state-of-the-art models, validating the robustness and efficiency of our proposed approach., Comment: This paper has been accepted by SIGIR 2024

Existing learning-based methods for solving job shop scheduling problems (JSSP) usually use off-the-shelf GNN models tailored to undirected graphs and neglect the rich and meaningful topological structures of disjunctive graphs (DGs). This paper proposes the topology-aware bidirectional graph attention network (TBGAT), a novel GNN architecture based on the attention mechanism, to embed the DG for solving JSSP in a local search framework. Specifically, TBGAT embeds the DG from a forward and a backward view, respectively, where the messages are propagated by following the different topologies of the views and aggregated via graph attention. Then, we propose a novel operator based on the message-passing mechanism to calculate the forward and backward topological sorts of the DG, which are the features for characterizing the topological structures and exploited by our model. In addition, we theoretically and experimentally show that TBGAT has linear computational complexity to the number of jobs and machines, respectively, strengthening our method's practical value. Besides, extensive experiments on five synthetic datasets and seven classic benchmarks show that TBGAT achieves new SOTA results by outperforming a wide range of neural methods by a large margin. All the code and data are publicly available online at https://github.com/zcaicaros/TBGAT.

While Large Language Models (LLMs) excel at the Winograd Schema Challenge (WSC), a coreference resolution task testing common-sense reasoning through pronoun disambiguation, they struggle with instances that feature minor alterations or rewording. To address this, we introduce EvoGrad, an open-source platform that harnesses a human-in-the-loop approach to create a dynamic dataset tailored to such altered WSC instances. Leveraging ChatGPT's capabilities, we expand our task instances from 182 to 3,691, setting a new benchmark for diverse common-sense reasoning datasets. Additionally, we introduce the error depth metric, assessing model stability in dynamic tasks. Our results emphasize the challenge posed by EvoGrad: Even the best performing LLM, GPT-3.5, achieves an accuracy of 65.0% with an average error depth of 7.2, a stark contrast to human performance of 92. 8% accuracy without perturbation errors. This highlights ongoing model limitations and the value of dynamic datasets in uncovering them., Comment: Accepted for publication in main proceedings of LREC-COLING 2024, 16 pages, 3 figures

Van der Waals (vdW) assembly could efficiently modulate the symmetry of two-dimensional (2D) materials that ultimately governs their physical properties. Of particular interest is the ferroelectric polarization being introduced by proper vdW assembly that enables the realization of novel electronic, magnetic and transport properties of 2D materials. Four-layer antiferromagnetic MnBi2Te4 (F-MBT) offers an excellent platform to explore ferroelectric polarization effects on magnetic order and topological transport properties of nanomaterials. Here, by applying symmetry analyses and density-functional-theory calculations, the ferroelectric interface effects on magnetic order, anomalous Hall effect (AHE) or even quantum AHE (QAHE) on the F-MBT are analyzed. Interlayer ferroelectric polarization in F-MBT efficiently violates the PT symmetry (the combination symmetry of central inversion (P) and time reverse (T) of the F-MBT by conferring magnetoelectric couplings, and stabilizes a specific antiferromagnetic order encompassing a ferromagnetic interface in the F-MBT. We predict that engineering an interlayer polarization in the top or bottom interface of F-MBT allows converting F-MBT from a trivial insulator to a Chern insulator. The switching of ferroelectric polarization at the middle interfaces results in a direction reversal of the quantum anomalous Hall current. Additionally, the interlayer polarization of the top and bottom interfaces can be aligned in the same direction, and the switching of polarization direction also reverses the direction of anomalous Hall currents. Overall, our work highlights the occurrence of quantum-transport phenomena in 2D vdW four-layer antiferromagnets through vdW assembly. These phenomena are absent in the bulk or thin-film in bulk-like stacking forms of MnBi2Te4.

We propose the weak magnetic effect, which emerges in quark-gluon plasma close to local thermal equilibrium as the dissipative correction to the quark phase space distribution function, as a novel contribution to the observed Lambda hyperon local spin polarization. With a finite field strength, which is consistent with previous estimate of the magnetic field in heavy-ion collisions, one is able to explain the experimentally observed Lambda local spin polarization through all centrality classes. Moreover, the weak magnetic effect plays an unambiguous role in the ordering between the second-order and the third-order modulations of the Lambda local spin polarization in experiments., Comment: Typos corrected

State of health (SOH) estimation for lithium-ion battery modules with cells connected in parallel is a challenging problem, especially with cell-to-cell variations. Incremental capacity analysis (ICA) and differential voltage analysis (DVA) are effective at the cell level, but a generalizable method to extend them to module-level SOH estimation remains missing, when only module-level measurements are available. This paper proposes a new method and demonstrates that, with multiple features systematically selected from the module-level ICA and DVA, the module-level SOH can be estimated with high accuracy and confidence in the presence of cell-to-cell variations. First, an information theory-based feature selection algorithm is proposed to find an optimal set of features for module-level SOH estimation. Second, a relevance vector regression (RVR)-based module-level SOH estimation model is proposed to provide both point estimates and three-sigma credible intervals while maintaining model sparsity. With more selected features incorporated, the proposed method achieves better estimation accuracy and higher confidence at the expense of higher model complexity. When applied to a large experimental dataset, the proposed method and the resulting sparse model lead to module-level SOH estimates with a 0.5% root-mean-square error and a 1.5% average three-sigma value. With all the training processes completed offboard, the proposed method has low computational complexity for onboard implementations., Comment: Addressed reviewer comments: Combined two sections, revised dataset and module-level result sections, corrected a typo in Algorithm 2; Previous Edit Comments: Condensed abstract; Added details in Introduction, Dataset, Module-Level Result Sections; Revised Section I, III & VII, IX; Added the initialization of Phi in Algorithm 2

There must be electromagnetic fields created during high-energy heavy-ion collisions. As the quark-gluon plasma (QGP) starts to evolve hydrodynamically, although these fields may become weak comparing to the energy scales of the strong interaction, they are potentially important to some electromagnetic probes. In this work, we focus on the dissipative corrections in QGP due to the presence of a weak external magnetic field, and calculate accordingly the induced photon radiation in the framework of viscous hydrodynamics. By event-by-event hydrodynamical simulations, the experimentally measured direct photon elliptic flow can be well reproduced. Correspondingly, the direct photon elliptic flow implies a magnetic field strength around 0.1$m_\pi^2 \sim 10^{16}$ G. This is indeed a weak field in heavy-ion physics that is compatible to the theoretical predictions, however, it is still an ultra-strong magnetic field in nature., Comment: version accepted by PRC

The identification of essential proteins can help in understanding the minimum requirements for cell survival and development. Network-based centrality approaches are commonly used to identify essential proteins from protein-protein interaction networks (PINs). Unfortunately, these approaches are limited by the poor quality of the underlying PIN data. To overcome this problem, researchers have focused on the prediction of essential proteins by combining PINs with other biological data. In this paper, we proposed a network refinement method based on module discovery and biological information to obtain a higher quality PIN. First, to extract the maximal connected subgraph in the PIN and to divide it into different modules by using Fast-unfolding algorithm; then, to detect critical modules based on the homology information, subcellular localization information and topology information within each module, and to construct a more refined network (CM-PIN). To evaluate the effectiveness of the proposed method, we used 10 typical network-based centrality methods (LAC, DC, DMNC, NC, TP, LID, CC, BC, PR, LR) to compare the overall performance of the CM-PIN with those the refined dynamic protein network (RD-PIN). The experimental results showed that the CM-PIN was optimal in terms of precision-recall curve, jackknife curve and other criteria, and can help to identify essential proteins more accurately.

Blockchain smart contracts have emerged as a transformative force in the digital realm, spawning a diverse range of compelling applications. Since solidity smart contracts across various domains manage trillions of dollars in virtual coins, they become a prime target for attacks. One of the primary challenges is keeping abreast of the latest techniques and tools for developing secure smart contracts and examining those already deployed. In this paper, we seek to address these challenges from four aspects: (1) We begin by examining ten automatic tools, specifically focusing on their methodologies and their ability to identify vulnerabilities in solidity smart contracts. (2) We propose a novel criterion for evaluating these tools, based on the ISO/IEC 25010 standard. (3) To facilitate the evaluation of the selected tools, we construct a benchmark that encompasses two distinct datasets: a collection of 389 labelled smart contracts and a scaled set of 20,000 unique cases from real-world contracts. (4) We provide a comparison of the selected tools, offering insights into their strengths and weaknesses and highlighting areas where further improvements are needed. Through this evaluation, we hope to provide developers and researchers with valuable guidance on selecting and using smart contract analysis tools and contribute to the ongoing efforts to improve the security and reliability of smart contracts., Comment: 24 pages, 6 figure, IEEE Communications Surveys & Tutorials

With the increasing adoption of smart contracts, ensuring their security has become a critical concern. Numerous vulnerabilities and attacks have been identified and exploited, resulting in significant financial losses. In response, researchers have developed various tools and techniques to identify and prevent vulnerabilities in smart contracts. In this survey, we present a systematic overview of the quality assurance of smart contracts, covering vulnerabilities, attacks, defenses, and tool support. By classifying vulnerabilities based on known attacks, we can identify patterns and common weaknesses that need to be addressed. Moreover, in order to effectively protect smart contracts, we have created a labeled dataset to evaluate various vulnerability detection tools and compare their effectiveness., Comment: 35 pages, 6 figures, ACM Computing Surveys

This paper explores the synergies between integrated power and thermal management (iPTM) and battery charging in an electric vehicle (EV). A multi-objective model predictive control (MPC) framework is developed to optimize the fast charging performance while enforcing the constraints in the power and thermal loops. The approach takes into account the coupling of the battery and cabin thermal management. The case study of a commercial EV demonstrates that the proposed method can effectively meet the requirements of fast charging and thermal management when accurate preview information is available. However, failure to predict the charging event can result in performance degradation with longer charging time. A time-varying weighting strategy is proposed to enhance charging performance in the presence of uncertainty. This strategy leverages the battery state-of-charge (SOC) and adjusts the priority of the multi-objective MPC at different phases during charging. Simulated results using a commercial EV use case show improved robustness in charging time using the proposed strategy., Comment: The 62nd Conference on Decision and Control (CDC), December 13-15, 2023, Singapore