Search

Cognitive diagnosis (CD) utilizes students' existing studying records to estimate their mastery of unknown knowledge concepts, which is vital for evaluating their learning abilities. Accurate CD is extremely challenging because CD is associated with complex relationships and mechanisms among students, knowledge concepts, studying records, etc. However, existing approaches loosely consider these relationships and mechanisms by a non-end-to-end learning framework, resulting in sub-optimal feature extractions and fusions for CD. Different from them, this paper innovatively proposes an End-to-end Graph Neural Networks-based Cognitive Diagnosis (EGNN-CD) model. EGNN-CD consists of three main parts: knowledge concept network (KCN), graph neural networks-based feature extraction (GNNFE), and cognitive ability prediction (CAP). First, KCN constructs CD-related interaction by comprehensively extracting physical information from students, exercises, and knowledge concepts. Second, a four-channel GNNFE is designed to extract high-order and individual features from the constructed KCN. Finally, CAP employs a multi-layer perceptron to fuse the extracted features to predict students' learning abilities in an end-to-end learning way. With such designs, the feature extractions and fusions are guaranteed to be comprehensive and optimal for CD. Extensive experiments on three real datasets demonstrate that our EGNN-CD achieves significantly higher accuracy than state-of-the-art models in CD.

A number of experiments show that the decay lifetimes of muons bound to atomic nuclei are longer than the decay lifetimes of free muons. In this paper, a scheme of extending quantum mechanics (EQM) is proposed to resolve this problem. The Schr$\ddot{\text{o}}$dinger's equation is obtained to prove the validation of this attempt. The decay ratio of bound muons is also calculated in EQM, and the result is in good agreement with the experimental data., Comment: 5 pages, 1 figure, 2 tables

Intelligent transportation systems (ITS) localization is of significant importance as it provides fundamental position and orientation for autonomous operations like intelligent vehicles. Integrating diverse and complementary sensors such as global navigation satellite system (GNSS) and 4D-radar can provide scalable and reliable global localization. Nevertheless, multi-sensor fusion encounters challenges including heterogeneity and time-varying uncertainty in measurements. Consequently, developing a reliable and unified multi-sensor framework remains challenging. In this paper, we introduce UniMSF, a comprehensive multi-sensor fusion localization framework for ITS, utilizing factor graphs. By integrating a multi-sensor fusion front-end, alongside outlier detection\&noise model estimation, and a factor graph optimization back-end, this framework accomplishes efficient fusion and ensures accurate localization for ITS. Specifically, in the multi-sensor fusion front-end module, we tackle the measurement heterogeneity among different modality sensors and establish effective measurement models. Reliable outlier detection and data-driven online noise estimation methods ensure that back-end optimization is immune to interference from outlier measurements. In addition, integrating multi-sensor observations via factor graph optimization offers the advantage of \enquote{plug and play}. Notably, our framework features high modularity and is seamlessly adapted to various sensor configurations. We demonstrate the effectiveness of the proposed framework through real vehicle tests by tightly integrating GNSS pseudorange and carrier phase information with IMU, and 4D-radar.

Its crux lies in the optimization of a tradeoff between accuracy and fairness of resultant models on the selected feature subset. The technical challenge of our setting is twofold: 1) streaming feature inputs, such that an informative feature may become obsolete or redundant for prediction if its information has been covered by other similar features that arrived prior to it, and 2) non-associational feature correlation, such that bias may be leaked from those seemingly admissible, non-protected features. To overcome this, we propose Streaming Feature Selection with Causal Fairness (SFCF) that builds two causal graphs egocentric to prediction label and protected feature, respectively, striving to model the complex correlation structure among streaming features, labels, and protected information. As such, bias can be eradicated from predictive modeling by removing those features being causally correlated with the protected feature yet independent to the labels. We theorize that the originally redundant features for prediction can later become admissible, when the learning accuracy is compromised by the large number of removed features (non-protected but can be used to reconstruct bias information). We benchmark SFCF\ on five datasets widely used in streaming feature research, and the results substantiate its performance superiority over six rival models in terms of efficiency and sparsity of feature selection and equalized odds of the resultant predictive models., Comment: This paper has been accepted by the 2024 IEEE International Conference on Systems, Man, and Cybernetics (SMC 2024)

Quantifying uncertainties for machine learning models is a critical step to reduce human verification effort by detecting predictions with low confidence. This paper proposes a method for uncertainty quantification (UQ) of table structure recognition (TSR). The proposed UQ method is built upon a mixture-of-expert approach termed Test-Time Augmentation (TTA). Our key idea is to enrich and diversify the table representations, to spotlight the cells with high recognition uncertainties. To evaluate the effectiveness, we proposed two heuristics to differentiate highly uncertain cells from normal cells, namely, masking and cell complexity quantification. Masking involves varying the pixel intensity to deem the detection uncertainty. Cell complexity quantification gauges the uncertainty of each cell by its topological relation with neighboring cells. The evaluation results based on standard benchmark datasets demonstrate that the proposed method is effective in quantifying uncertainty in TSR models. To our best knowledge, this study is the first of its kind to enable UQ in TSR tasks. Our code and data are available at: https://github.com/lamps-lab/UQTTA.git., Comment: 7 Figures

Bluetooth Low Energy (BLE) is a short-range wireless communication technology for resource-constrained IoT devices. Unfortunately, BLE is vulnerable to session-based attacks, where previous packets construct exploitable conditions for subsequent packets to compromise connections. Defending against session-based attacks is challenging because each step in the attack sequence is legitimate when inspected individually. In this paper, we present BlueSWAT, a lightweight state-aware security framework for protecting BLE devices. To perform inspection on the session level rather than individual packets, BlueSWAT leverages a finite state machine (FSM) to monitor sequential actions of connections at runtime. Patterns of session-based attacks are modeled as malicious transition paths in the FSM. To overcome the heterogeneous IoT environment, we develop a lightweight eBPF framework to facilitate universal patch distribution across different BLE architectures and stacks, without requiring device reboot. We implement BlueSWAT on 5 real-world devices with different chips and stacks to demonstrate its cross-device adaptability. On our dataset with 101 real-world BLE vulnerabilities, BlueSWAT can mitigate 76.1% of session-based attacks, outperforming other defense frameworks. In our end-to-end application evaluation, BlueSWAT patches introduce an average of 0.073% memory overhead and negligible latency.

The stellar-to-halo mass relation (SHMR) is a fundamental relationship between galaxies and their host dark matter haloes. In this study, we examine the scatter in this relation for primary galaxies in the semi-analytic L-Galaxies model and two cosmological hydrodynamical simulations, \eagle{} and \tng{}. We find that in low-mass haloes, more massive galaxies tend to reside in haloes with higher concentration, earlier formation time, greater environmental density, earlier major mergers, and, to have older stellar populations, which is consistent with findings in various studies. Quantitative analysis reveals the varying significance of halo and galaxy properties in determining SHMR scatter across simulations and models. In \eagle{} and \tng{}, halo concentration and formation time primarily influence SHMR scatter for haloes with $M_{\rm h}<10^{12}~\rm M_\odot$, but the influence diminishes at high mass. Baryonic processes play a more significant role in \lgal{}. For halos with $M_{\rm h} <10^{11}~\rm M_\odot$ and $10^{12}~\rm M_\odot 10^{13}~\rm M_\odot$, supermassive black hole mass becomes more important. Interestingly, it is found that AGN feedback may increase the amplitude of the scatter and decrease the dependence on halo properties at high masses., Comment: 23 pages, 12 + 5 figures, 2 tables, including 4 Appendix; Accepted by MNRAS

System services and resources in Android are accessed through IPC based mechanisms. Previous research has demonstrated that they are vulnerable to the denial-of-service attack (DoS attack). For instance, the JNI global reference (JGR), which is widely used by system services, can be exhausted to cause the system reboot (hence the name JGRE attack). Even though the Android team tries to fix the problem by enforcing security checks, we find that it is still possible to construct a JGR exhaustion DoS attack in the latest Android system. In this paper, we propose a new JGR exhaustion DoS attack, which is effective in different Android versions, including the latest one (i.e., Android 10). Specifically, we developed JGREAnalyzer, a tool that can systematically detect JGR vulnerable services APIs via a call graph analysis and a forwarding reachability analysis. We applied this tool to different Android versions and found multiple vulnerabilities. In particular, among 148 system services in Android 10, 12 of them have 21 vulnerabilities. Among them, 9 can be successfully exploited without any permissions. We further analyze the root cause of the vulnerabilities and propose a new defense to mitigate the JGRE attack by restricting resource consumption via global reference counting.

Creating high-fidelity head avatars from multi-view videos is a core issue for many AR/VR applications. However, existing methods usually struggle to obtain high-quality renderings for all different head components simultaneously since they use one single representation to model components with drastically different characteristics (e.g., skin vs. hair). In this paper, we propose a Hybrid Mesh-Gaussian Head Avatar (MeGA) that models different head components with more suitable representations. Specifically, we select an enhanced FLAME mesh as our facial representation and predict a UV displacement map to provide per-vertex offsets for improved personalized geometric details. To achieve photorealistic renderings, we obtain facial colors using deferred neural rendering and disentangle neural textures into three meaningful parts. For hair modeling, we first build a static canonical hair using 3D Gaussian Splatting. A rigid transformation and an MLP-based deformation field are further applied to handle complex dynamic expressions. Combined with our occlusion-aware blending, MeGA generates higher-fidelity renderings for the whole head and naturally supports more downstream tasks. Experiments on the NeRSemble dataset demonstrate the effectiveness of our designs, outperforming previous state-of-the-art methods and supporting various editing functionalities, including hairstyle alteration and texture editing., Comment: Project page: https://conallwang.github.io/MeGA_Pages/

Recent large vision models (e.g., SAM) enjoy great potential to facilitate intelligent perception with high accuracy. Yet, the resource constraints in the IoT environment tend to limit such large vision models to be locally deployed, incurring considerable inference latency thereby making it difficult to support real-time applications, such as autonomous driving and robotics. Edge-cloud collaboration with large-small model co-inference offers a promising approach to achieving high inference accuracy and low latency. However, existing edge-cloud collaboration methods are tightly coupled with the model architecture and cannot adapt to the dynamic data drifts in heterogeneous IoT environments. To address the issues, we propose LAECIPS, a new edge-cloud collaboration framework. In LAECIPS, both the large vision model on the cloud and the lightweight model on the edge are plug-and-play. We design an edge-cloud collaboration strategy based on hard input mining, optimized for both high accuracy and low latency. We propose to update the edge model and its collaboration strategy with the cloud under the supervision of the large vision model, so as to adapt to the dynamic IoT data streams. Theoretical analysis of LAECIPS proves its feasibility. Experiments conducted in a robotic semantic segmentation system using real-world datasets show that LAECIPS outperforms its state-of-the-art competitors in accuracy, latency, and communication overhead while having better adaptability to dynamic environments.

The proliferation of consumer IoT products in our daily lives has raised the need for secure device authentication and access control. Unfortunately, these resource-constrained devices typically use token-based authentication, which is vulnerable to token compromise attacks that allow attackers to impersonate the devices and perform malicious operations by stealing the access token. Using hardware fingerprints to secure their authentication is a promising way to mitigate these threats. However, once attackers have stolen some hardware fingerprints (e.g., via MitM attacks), they can bypass the hardware authentication by training a machine learning model to mimic fingerprints or reusing these fingerprints to craft forge requests. In this paper, we present MCU-Token, a secure hardware fingerprinting framework for MCU-based IoT devices even if the cryptographic mechanisms (e.g., private keys) are compromised. MCU-Token can be easily integrated with various IoT devices by simply adding a short hardware fingerprint-based token to the existing payload. To prevent the reuse of this token, we propose a message mapping approach that binds the token to a specific request via generating the hardware fingerprints based on the request payload. To defeat the machine learning attacks, we mix the valid fingerprints with poisoning data so that attackers cannot train a usable model with the leaked tokens. MCU-Token can defend against armored adversary who may replay, craft, and offload the requests via MitM or use both hardware (e.g., use identical devices) and software (e.g., machine learning attacks) strategies to mimic the fingerprints. The system evaluation shows that MCU-Token can achieve high accuracy (over 97%) with a low overhead across various IoT devices and application scenarios.

Face images contain a wide variety of attribute information. In this paper, we propose a generalized framework for joint estimation of ordinal and nominal attributes based on information sharing. We tackle the correlation problem between heterogeneous attributes using hard parameter sharing of shallow features, and trade-off multiple loss functions by considering homoskedastic uncertainty for each attribute estimation task. This leads to optimal estimation of multiple attributes of the face and reduces the training cost of multitask learning. Experimental results on benchmarks with multiple face attributes show that the proposed approach has superior performance compared to state of the art. Finally, we discuss the bias issues arising from the proposed approach in face attribute estimation and validate its feasibility on edge systems.

Smart home IoT systems utilize trigger-action platforms, e.g., IFTTT, to manage devices from various vendors. However, they may be abused by triggering malicious rule execution with forged IoT devices or events violating the execution integrity and the intentions of the users. To address this issue, we propose a ledger based IoT platform called Ruledger, which ensures the correct execution of rules by verifying the authenticity of the corresponding information. Ruledger utilizes smart contracts to enforce verifying the information associated with rule executions, e.g., the user and configuration information from users, device events, and triggers in the trigger-action platforms. In particular, we develop three algorithms to enable ledger-wallet based applications for Ruledger and guarantee that the records used for verification are stateful and correct. Thus, the execution integrity of rules is ensured even if devices and platforms in the smart home systems are compromised. We prototype Ruledger in a real IoT platform, i.e., IFTTT, and evaluate the performance with various settings. The experimental results demonstrate Ruledger incurs an average of 12.53% delay, which is acceptable for smart home systems.

Narrative visualization effectively transforms data into engaging stories, making complex information accessible to a broad audience. Foundation models, essential for narrative visualization, inherently facilitate this process through their superior ability to handle natural language queries and answers, generate cohesive narratives, and enhance visual communication. Inspired by previous work in narrative visualization and recent advances in foundation models, we synthesized potential tasks and opportunities for foundation models at various stages of narrative visualization. In our study, we surveyed 77 papers to explore the role of foundation models in automatingnarrative visualization creation. We propose a reference model that leverages foundation models for crafting narrative visualization, categorizing the reviewed literature into four essential phases: Analysis, Narration, Visualization, and Interaction. Additionally, we identifyeight specific tasks where foundation models are applied across these stages. This study maps out the landscape of challenges and opportunities, providing insightful directions for future research and valuable guidance for scholars in the field., Comment: 20 pages,6 figures, 2 tables

Infrared and Raman spectroscopy are widely used for the characterization of gases, liquids, and solids, as the spectra contain a wealth of information concerning in particular the dynamics of these systems. Atomic scale simulations can be used to predict such spectra but are often severely limited due to high computational cost or the need for strong approximations that limit application range and reliability. Here, we introduce a machine learning (ML) accelerated approach that addresses these shortcomings and provides a significant performance boost in terms of data and computational efficiency compared to earlier ML schemes. To this end, we generalize the neuroevolution potential approach to enable the prediction of rank one and two tensors to obtain the tensorial neuroevolution potential (TNEP) scheme. We apply the resulting framework to construct models for the dipole moment, polarizability, and susceptibility of molecules, liquids, and solids, and show that our approach compares favorably with several ML models from the literature with respect to accuracy and computational efficiency. Finally, we demonstrate the application of the TNEP approach to the prediction of infrared and Raman spectra of liquid water, a molecule (PTAF-), and a prototypical perovskite with strong anharmonicity (BaZrO3). The TNEP approach is implemented in the free and open source software package GPUMD, which makes this methodology readily available to the scientific community., Comment: 11 pages, 6 figures

QUIC is expected to be a game-changer in improving web application performance. In this paper, we conduct a systematic examination of QUIC's performance over high-speed networks. We find that over fast Internet, the UDP+QUIC+HTTP/3 stack suffers a data rate reduction of up to 45.2% compared to the TCP+TLS+HTTP/2 counterpart. Moreover, the performance gap between QUIC and HTTP/2 grows as the underlying bandwidth increases. We observe this issue on lightweight data transfer clients and major web browsers (Chrome, Edge, Firefox, Opera), on different hosts (desktop, mobile), and over diverse networks (wired broadband, cellular). It affects not only file transfers, but also various applications such as video streaming (up to 9.8% video bitrate reduction) and web browsing. Through rigorous packet trace analysis and kernel- and user-space profiling, we identify the root cause to be high receiver-side processing overhead, in particular, excessive data packets and QUIC's user-space ACKs. We make concrete recommendations for mitigating the observed performance issues., Comment: 10 pages, 16 figures

Document AI aims to automatically analyze documents by leveraging natural language processing and computer vision techniques. One of the major tasks of Document AI is document layout analysis, which structures document pages by interpreting the content and spatial relationships of layout, image, and text. This task can be image-centric, wherein the aim is to identify and label various regions such as authors and paragraphs, or text-centric, where the focus is on classifying individual words in a document. Although there are increasingly sophisticated methods for improving layout analysis, doubts remain about the extent to which their findings can be generalized to a broader context. Specifically, prior work developed systems based on very different architectures, such as transformer-based, graph-based, and CNNs. However, no work has mentioned the effectiveness of these models in a comparative analysis. Moreover, while language-independent Document AI models capable of knowledge transfer have been developed, it remains to be investigated to what degree they can effectively transfer knowledge. In this study, we aim to fill these gaps by conducting a comparative evaluation of state-of-the-art models in document layout analysis and investigating the potential of cross-lingual layout analysis by utilizing machine translation techniques.

End-to-end (E2E) systems have shown comparable performance to hybrid systems for automatic speech recognition (ASR). Word timings, as a by-product of ASR, are essential in many applications, especially for subtitling and computer-aided pronunciation training. In this paper, we improve the frame-level classifier for word timings in E2E system by introducing label priors in connectionist temporal classification (CTC) loss, which is adopted from prior works, and combining low-level Mel-scale filter banks with high-level ASR encoder output as input feature. On the internal Chinese corpus, the proposed method achieves 95.68%/94.18% compared to the hybrid system 93.0%/90.22% on the word timing accuracy metrics. It also surpass a previous E2E approach with an absolute increase of 4.80%/8.02% on the metrics on 7 languages. In addition, we further improve word timing accuracy by delaying CTC peaks with frame-wise knowledge distillation, though only experimenting on LibriSpeech., Comment: To appear in the proceedings of INTERSPEECH 2023

Deep forest is a non-differentiable deep model which has achieved impressive empirical success across a wide variety of applications, especially on categorical/symbolic or mixed modeling tasks. Many of the application fields prefer explainable models, such as random forests with feature contributions that can provide local explanation for each prediction, and Mean Decrease Impurity (MDI) that can provide global feature importance. However, deep forest, as a cascade of random forests, possesses interpretability only at the first layer. From the second layer on, many of the tree splits occur on the new features generated by the previous layer, which makes existing explanatory tools for random forests inapplicable. To disclose the impact of the original features in the deep layers, we design a calculation method with an estimation step followed by a calibration step for each layer, and propose our feature contribution and MDI feature importance calculation tools for deep forest. Experimental results on both simulated data and real world data verify the effectiveness of our methods.

3D model reconstruction from a single image has achieved great progress with the recent deep generative models. However, the conventional reconstruction approaches with template mesh deformation and implicit fields have difficulty in reconstructing non-watertight 3D mesh models, such as garments. In contrast to image-based modeling, the sketch-based approach can help users generate 3D models to meet the design intentions from hand-drawn sketches. In this study, we propose Sketch2Cloth, a sketch-based 3D garment generation system using the unsigned distance fields from the user's sketch input. Sketch2Cloth first estimates the unsigned distance function of the target 3D model from the sketch input, and extracts the mesh from the estimated field with Marching Cubes. We also provide the model editing function to modify the generated mesh. We verified the proposed Sketch2Cloth with quantitative evaluations on garment generation and editing with a state-of-the-art approach., Comment: 8 pages, 9 figures, video is here https://youtu.be/miisvVTpqj8

The increasing use of Machine Learning (ML) software can lead to unfair and unethical decisions, thus fairness bugs in software are becoming a growing concern. Addressing these fairness bugs often involves sacrificing ML performance, such as accuracy. To address this issue, we present a novel counterfactual approach that uses counterfactual thinking to tackle the root causes of bias in ML software. In addition, our approach combines models optimized for both performance and fairness, resulting in an optimal solution in both aspects. We conducted a thorough evaluation of our approach on 10 benchmark tasks using a combination of 5 performance metrics, 3 fairness metrics, and 15 measurement scenarios, all applied to 8 real-world datasets. The conducted extensive evaluations show that the proposed method significantly improves the fairness of ML software while maintaining competitive performance, outperforming state-of-the-art solutions in 84.6% of overall cases based on a recent benchmarking tool.

Polarizability is closely related to many fundamental characteristics of molecular systems and plays an indispensable role in simulating the Raman spectra. However, the calculations of polarizability for large systems still suffers from the limitations of processing ability of the quantum mechanical (QM) methods. This work assessed and compared the accuracy of the bond polarizability model (BPM) and a ML-based atomic polarizability model (AlphaML) in predicting polarizability of alkanes and then also investigated the ability of simulating Raman spectra. We found that the AlphaML has appreciable advantages over the BPM in learning the polarizability in the training data set and predicting polarizability of molecules that configurational differently from training structures. In addition, the BPM has inherent disadvantages in predicting polarizability anisotropy due to many factors including large uncertainties of estimating bond anisotropy, omitting of off-diagonal parameters in the construction of the model. As a result, the BPM has larger errors than the AlphaML in the simulation of anisotropic Raman scattering. Finally, we demonstrated that both the BPM and AlphaML suffer from transference to alkanes larger than those used in the training data sets, but the problem for the AlphaML can be circumvented by exploring more proper training structures., Comment: 37 pages, 10 figures

User behavior data produced during interaction with massive items in the significant data era are generally heterogeneous and sparse, leaving the recommender system (RS) a large diversity of underlying patterns to excavate. Deep neural network-based models have reached the state-of-the-art benchmark of the RS owing to their fitting capabilities. However, prior works mainly focus on designing an intricate architecture with fixed loss function and regulation. These single-metric models provide limited performance when facing heterogeneous and sparse user behavior data. Motivated by this finding, we propose a multi-metric AutoRec (MMA) based on the representative AutoRec. The idea of the proposed MMA is mainly two-fold: 1) apply different $L_p$-norm on loss function and regularization to form different variant models in different metric spaces, and 2) aggregate these variant models. Thus, the proposed MMA enjoys the multi-metric orientation from a set of dispersed metric spaces, achieving a comprehensive representation of user data. Theoretical studies proved that the proposed MMA could attain performance improvement. The extensive experiment on five real-world datasets proves that MMA can outperform seven other state-of-the-art models in predicting unobserved user behavior data., Comment: 6 pages, 4 Tables

Mid-term and long-term electric energy demand prediction is essential for the planning and operations of the smart grid system. Mainly in countries where the power system operates in a deregulated environment. Traditional forecasting models fail to incorporate external knowledge while modern data-driven ignore the interpretation of the model, and the load series can be influenced by many complex factors making it difficult to cope with the highly unstable and nonlinear power load series. To address the forecasting problem, we propose a more accurate district level load prediction model Based on domain knowledge and the idea of decomposition and ensemble. Its main idea is three-fold: a) According to the non-stationary characteristics of load time series with obvious cyclicality and periodicity, decompose into series with actual economic meaning and then carry out load analysis and forecast. 2) Kernel Principal Component Analysis(KPCA) is applied to extract the principal components of the weather and calendar rule feature sets to realize data dimensionality reduction. 3) Give full play to the advantages of various models based on the domain knowledge and propose a hybrid model(XASXG) based on Autoregressive Integrated Moving Average model(ARIMA), support vector regression(SVR) and Extreme gradient boosting model(XGBoost). With such designs, it accurately forecasts the electricity demand in spite of their highly unstable characteristic. We compared our method with nine benchmark methods, including classical statistical models as well as state-of-the-art models based on machine learning, on the real time series of monthly electricity demand in four Chinese cities. The empirical study shows that the proposed hybrid model is superior to all competitors in terms of accuracy and prediction bias.

One of limitations in end-to-end automatic speech recognition (ASR) framework is its performance would be compromised if train-test utterance lengths are mismatched. In this paper, we propose an on-the-fly random utterance concatenation (RUC) based data augmentation method to alleviate train-test utterance length mismatch issue for short-video ASR task. Specifically, we are motivated by observations that our human-transcribed training utterances tend to be much shorter for short-video spontaneous speech (~3 seconds on average), while our test utterance generated from voice activity detection front-end is much longer (~10 seconds on average). Such a mismatch can lead to suboptimal performance. Empirically, it's observed the proposed RUC method significantly improves long utterance recognition without performance drop on short one. Overall, it achieves 5.72% word error rate reduction on average for 15 languages and improved robustness to various utterance length., Comment: 5 pages, 3 figures, 4 tables

Code-switching (CS) refers to the phenomenon that languages switch within a speech signal and leads to language confusion for automatic speech recognition (ASR). This paper aims to address language confusion for improving CS-ASR from two perspectives: incorporating and disentangling language information. We incorporate language information in the CS-ASR model by dynamically biasing the model with token-level language posteriors which are outputs of a sequence-to-sequence auxiliary language diarization module. In contrast, the disentangling process reduces the difference between languages via adversarial training so as to normalize two languages. We conduct the experiments on the SEAME dataset. Compared to the baseline model, both the joint optimization with LD and the language posterior bias achieve performance improvement. The comparison of the proposed methods indicates that incorporating language information is more effective than disentangling for reducing language confusion in CS speech., Comment: Submitted to ICASSP 2023

Light-sheet fluorescence microscopy (LSFM) makes use of a thin plane of light to optically section and image transparent tissues or organisms {\it{in vivo}}, which has the advantages of fast imaging speed and low phototoxicity. In this paper, we have employed light-sheet microscopy to investigate the polarization effects on fluorescence emission of zebrafish neurons via modifying the electric oscillation orientation of the excitation light. The intensity of the fluorescence emission from the excited zebrafish larvae follows a cosine square function with respect to the polarization state of the excitation light and reveals a 40$\%$ higher fluorescence emission when the polarization orientation is orthogonal to the illumination and detection axes. Through registration and subtraction of fluorescence images under different polarization states, we have demonstrated that most of the enhanced fluorescence signals are from the nerve cells rather than the extracellular substance. This provides us a way to distinguish the cell boundaries and observe the organism structures with improved contrast and resolution.

We present the NLO electroweak radiative corrections to the ee\gamma production in \gamma\gamma collision, which is an ideal channel for calibrating the beam luminosity of a Photon Linear Collider. We analyze the dependence of the total cross section on the beam colliding energy, and then investigate the kinematic distributions of final particles at various initial photon beam polarizations at EW NLO accuracy. The numerical results indicate that the EW relative corrections to the total cross section are non-negligible and become increasingly significant as the increase of the beam colliding energy. Such EW corrections are very important and should be taken into consideration in precision theoretical and experimental studies at high-energy \gamma\gamma colliders., Comment: 32 pages, 50 figures, 3 tables

Online streaming feature selection (OSFS), which conducts feature selection in an online manner, plays an important role in dealing with high-dimensional data. In many real applications such as intelligent healthcare platform, streaming feature always has some missing data, which raises a crucial challenge in conducting OSFS, i.e., how to establish the uncertain relationship between sparse streaming features and labels. Unfortunately, existing OSFS algorithms never consider such uncertain relationship. To fill this gap, we in this paper propose an online sparse streaming feature selection with uncertainty (OS2FSU) algorithm. OS2FSU consists of two main parts: 1) latent factor analysis is utilized to pre-estimate the missing data in sparse streaming features before con-ducting feature selection, and 2) fuzzy logic and neighborhood rough set are employed to alleviate the uncertainty between estimated streaming features and labels during conducting feature selection. In the experiments, OS2FSU is compared with five state-of-the-art OSFS algorithms on six real datasets. The results demonstrate that OS2FSU outperforms its competitors when missing data are encountered in OSFS.

We consider the following singularly perturbed elliptic problem \[ - {\varepsilon ^2}\Delta u + u = f(u){\text{ in }}\Omega ,{\text{ }}u > 0{\text{ in }}\Omega ,{\text{ }}u = 0{\text{ on }}\partial \Omega , \] where $\Omega$ is a domain in ${\mathbb{R}^N}(N \ge 3)$, not necessarily bounded, with boundary $\partial \Omega \in {C^2}$ and the nonlinearity $f$ is of critical growth. In this paper, we construct a family of multi-peak solutions to the equation given above which concentrate around any prescribed finite sets of local maxima of the distance function from the boundary $\partial \Omega$., Comment: 42 pages

Internal Language Model Estimation (ILME) based language model (LM) fusion has been shown significantly improved recognition results over conventional shallow fusion in both intra-domain and cross-domain speech recognition tasks. In this paper, we attempt to apply our ILME method to cross-domain code-switching speech recognition (CSSR) work. Specifically, our curiosity comes from several aspects. First, we are curious about how effective the ILME-based LM fusion is for both intra-domain and cross-domain CSSR tasks. We verify this with or without merging two code-switching domains. More importantly, we train an end-to-end (E2E) speech recognition model by means of merging two monolingual data sets and observe the efficacy of the proposed ILME-based LM fusion for CSSR. Experimental results on SEAME that is from Southeast Asian and another Chinese Mainland CS data set demonstrate the effectiveness of the proposed ILME-based LM fusion method., Comment: 5 pages. Submitted to INTERSPEECH 2022

Intermediate layer output (ILO) regularization by means of multitask training on encoder side has been shown to be an effective approach to yielding improved results on a wide range of end-to-end ASR frameworks. In this paper, we propose a novel method to do ILO regularized training differently. Instead of using conventional multitask methods that entail more training overhead, we directly make the intermediate layer output as input to the decoder, that is, our decoder not only accepts the output of the final encoder layer as input, it also takes the output of the encoder ILO as input during training. With the proposed method, as both encoder and decoder are simultaneously "regularized", the network is more sufficiently trained, consistently leading to improved results, over the ILO-based CTC method, as well as over the original attention-based modeling method without the proposed method employed., Comment: 5 pages. Submitted to INTERSPEECH 2022

Attention guidance is an approach to addressing dataset bias in deep learning, where the model relies on incorrect features to make decisions. Focusing on image classification tasks, we propose an efficient human-in-the-loop system to interactively direct the attention of classifiers to the regions specified by users, thereby reducing the influence of co-occurrence bias and improving the transferability and interpretability of a DNN. Previous approaches for attention guidance require the preparation of pixel-level annotations and are not designed as interactive systems. We present a new interactive method to allow users to annotate images with simple clicks, and study a novel active learning strategy to significantly reduce the number of annotations. We conducted both a numerical evaluation and a user study to evaluate the proposed system on multiple datasets. Compared to the existing non-active-learning approach which usually relies on huge amounts of polygon-based segmentation masks to fine-tune or train the DNNs, our system can save lots of labor and money and obtain a fine-tuned network that works better even when the dataset is biased. The experiment results indicate that the proposed system is efficient, reasonable, and reliable., Comment: 13 pages, 11 figures, proceeding of ACM IUI 2023, video https://youtu.be/2MD-z6vXKJ4

This paper investigates a new online learning problem with doubly-streaming data, where the data streams are described by feature spaces that constantly evolve, with new features emerging and old features fading away. The challenges of this problem are two folds: 1) Data samples ceaselessly flowing in may carry shifted patterns over time, requiring learners to update hence adapt on-the-fly. 2) Newly emerging features are described by very few samples, resulting in weak learners that tend to make error predictions. A plausible idea to overcome the challenges is to establish relationship between the pre-and-post evolving feature spaces, so that an online learner can leverage the knowledge learned from the old features to better the learning performance on the new features. Unfortunately, this idea does not scale up to high-dimensional media streams with complex feature interplay, which suffers an tradeoff between onlineness (biasing shallow learners) and expressiveness(requiring deep learners). Motivated by this, we propose a novel OLD^3S paradigm, where a shared latent subspace is discovered to summarize information from the old and new feature spaces, building intermediate feature mapping relationship. A key trait of OLD^3S is to treat the model capacity as a learnable semantics, yields optimal model depth and parameters jointly, in accordance with the complexity and non-linearity of the input data streams in an online fashion. Both theoretical analyses and empirical studies substantiate the viability and effectiveness of our proposal., Comment: Have accepted by ACMMM 2022. Legends mistake in Figure 4 has been corrected

High-dimensional and sparse (HiDS) matrices are omnipresent in a variety of big data-related applications. Latent factor analysis (LFA) is a typical representation learning method that extracts useful yet latent knowledge from HiDS matrices via low-rank approximation. Current LFA-based models mainly focus on a single-metric representation, where the representation strategy designed for the approximation Loss function, is fixed and exclusive. However, real-world HiDS matrices are commonly heterogeneous and inclusive and have diverse underlying patterns, such that a single-metric representation is most likely to yield inferior performance. Motivated by this, we in this paper propose a multi-metric latent factor (MMLF) model. Its main idea is two-fold: 1) two vector spaces and three Lp-norms are simultaneously employed to develop six variants of LFA model, each of which resides in a unique metric representation space, and 2) all the variants are ensembled with a tailored, self-adaptive weighting strategy. As such, our proposed MMLF enjoys the merits originated from a set of disparate metric spaces all at once, achieving the comprehensive and unbiased representation of HiDS matrices. Theoretical study guarantees that MMLF attains a performance gain. Extensive experiments on eight real-world HiDS datasets, spanning a wide range of industrial and science domains, verify that our MMLF significantly outperforms ten state-of-the-art, shallow and deep counterparts.

A High-dimensional and sparse (HiDS) matrix is frequently encountered in a big data-related application like an e-commerce system or a social network services system. To perform highly accurate representation learning on it is of great significance owing to the great desire of extracting latent knowledge and patterns from it. Latent factor analysis (LFA), which represents an HiDS matrix by learning the low-rank embeddings based on its observed entries only, is one of the most effective and efficient approaches to this issue. However, most existing LFA-based models perform such embeddings on a HiDS matrix directly without exploiting its hidden graph structures, thereby resulting in accuracy loss. To address this issue, this paper proposes a graph-incorporated latent factor analysis (GLFA) model. It adopts two-fold ideas: 1) a graph is constructed for identifying the hidden high-order interaction (HOI) among nodes described by an HiDS matrix, and 2) a recurrent LFA structure is carefully designed with the incorporation of HOI, thereby improving the representa-tion learning ability of a resultant model. Experimental results on three real-world datasets demonstrate that GLFA outperforms six state-of-the-art models in predicting the missing data of an HiDS matrix, which evidently supports its strong representation learning ability to HiDS data.

Android allows apps to communicate with its system services via system service helpers so that these apps can use various functions provided by the system services. Meanwhile, the system services rely on their service helpers to enforce security checks for protection. Unfortunately, the security checks in the service helpers may be bypassed via directly exploiting the non-SDK (hidden) APIs, degrading the stability and posing severe security threats such as privilege escalation, automatic function execution without users' interactions, crashes, and DoS attacks. Google has proposed various approaches to address this problem, e.g., case-by-case fixing the bugs or even proposing a blacklist to block all the non-SDK APIs. However, the developers can still figure out new ways of exploiting these hidden APIs to evade the non-SDKs restrictions. In this paper, we systematically study the vulnerabilities due to the hidden API exploitation and analyze the effectiveness of Google's countermeasures. We aim to answer if there are still vulnerable hidden APIs that can be exploited in the newest Android 12. We develop a static analysis tool called ServiceAudit to automatically mine the inconsistent security enforcement between service helper classes and the hidden service APIs. We apply ServiceAudit to Android 6~12. Our tool discovers 112 vulnerabilities in Android 6 with higher precision than existing approaches. Moreover, in Android 11 and 12, we identify more than 25 hidden APIs with inconsistent protections; however, only one of the vulnerable APIs can lead to severe security problems in Android 11, and none of them work on Android 12.

An end-to-end (E2E) ASR model implicitly learns a prior Internal Language Model (ILM) from the training transcripts. To fuse an external LM using Bayes posterior theory, the log likelihood produced by the ILM has to be accurately estimated and subtracted. In this paper we propose two novel approaches to estimate the ILM based on Listen-Attend-Spell (LAS) framework. The first method is to replace the context vector of the LAS decoder at every time step with a vector that is learned with training transcripts. Furthermore, we propose another method that uses a lightweight feed-forward network to directly map query vector to context vector in a dynamic sense. Since the context vectors are learned by minimizing the perplexities on training transcripts, and their estimation is independent of encoder output, hence the ILMs are accurately learned for both methods. Experiments show that the ILMs achieve the lowest perplexity, indicating the efficacy of the proposed methods. In addition, they also significantly outperform the shallow fusion method, as well as two previously proposed ILM Estimation (ILME) approaches on several datasets., Comment: Proceedings of INTERSPEECH

3D modeling based on point clouds is an efficient way to reconstruct and create detailed 3D content. However, the geometric procedure may lose accuracy due to high redundancy and the absence of an explicit structure. In this work, we propose a human-in-the-loop sketch-based point cloud reconstruction framework to leverage users cognitive abilities in geometry extraction. We present an interactive drawing interface for 3D model creation from point cloud data with the help of user sketches. We adopt an optimization method in which the user can continuously edit the contours extracted from the obtained 3D model and retrieve the model iteratively. Finally, we verify the proposed user interface for modeling from sparse point clouds. see video here https://www.youtube.com/watch?v=0H19NyXDRJE ., Comment: 6 pages, 5 figures, accepted in Proceedings of International Workshop on Advanced Image Technology 2022

We thoroughly investigate both transverse momentum and threshold resummation effects on scalar-pseudoscalar pair production via quark-antiquark annihilation at the $13~ \text{TeV}$ Large Hadron Collider at QCD NLO+NLL accuracy. A factorization method is introduced to properly supplement the soft-gluon (threshold) resummation contribution from parton distribution functions to the resummed results obtained by the Collins-Soper-Sterman resummation approach. We find that the impact of the threshold-resummation improved PDFs is comparable to the resummation effect of the partonic matrix element and can even predominate in high invariant mass regions. Moreover, the loop-induced gluon-gluon fusion channel in the type-I two-Higgs-doublet model is considered in our calculation. The numerical results show that the electroweak production via quark-antiquark annihilation dominates over the gluon-initiated QCD production by $1 \sim 2$ orders of magnitude., Comment: 29 pages, 6 figures

Modeling 3D context is essential for high-performance 3D medical image analysis. Although 2D networks benefit from large-scale 2D supervised pretraining, it is weak in capturing 3D context. 3D networks are strong in 3D context yet lack supervised pretraining. As an emerging technique, \emph{3D context fusion operator}, which enables conversion from 2D pretrained networks, leverages the advantages of both and has achieved great success. Existing 3D context fusion operators are designed to be spatially symmetric, i.e., performing identical operations on each 2D slice like convolutions. However, these operators are not truly equivariant to translation, especially when only a few 3D slices are used as inputs. In this paper, we propose a novel asymmetric 3D context fusion operator (A3D), which uses different weights to fuse 3D context from different 2D slices. Notably, A3D is NOT translation-equivariant while it significantly outperforms existing symmetric context fusion operators without introducing large computational overhead. We validate the effectiveness of the proposed method by extensive experiments on DeepLesion benchmark, a large-scale public dataset for universal lesion detection from computed tomography (CT). The proposed A3D consistently outperforms symmetric context fusion operators by considerable margins, and establishes a new \emph{state of the art} on DeepLesion. To facilitate open research, our code and model in PyTorch are available at https://github.com/M3DV/AlignShift., Comment: MICCAI 2021. The code and model are available at https://github.com/M3DV/AlignShift

We study the approaches to two-dimensional integrable field theories via a six-dimensional(6D) holomorphic Chern-Simons theory defined on twistor space. Under symmetry reduction, it reduces to a four-dimensional Chern-Simons theory, while under solving along fibres it leads to four-dimensional(4D) integrable theory, the anti-self-dual Yang-Mills or its generalizations. From both four-dimensional theories, various two-dimensional integrable field theories can be obtained. In this work, we try to investigate several two-dimensional integrable deformations in this framework. We find that the $\lambda$-deformation, the rational $\eta$-deformation and the generalized $\lambda$-deformation can not be realized from 4D integrable model approach, even though they could be obtained from 4D Chern-Simons theory. The obstacle stems from the incompatibility between the symmetry reduction and the boundary conditions. Nevertheless, we show that a coupled theory of $\lambda$-deformation and the $\eta$-deformation in the trigonometric description could be obtained from the 6D theory in both ways, by considering the case that $(3,0)$-form in the 6D theory is allowed to have zeros.

Normal map is an important and efficient way to represent complex 3D models. A designer may benefit from the auto-generation of high quality and accurate normal maps from freehand sketches in 3D content creation. This paper proposes a deep generative model for generating normal maps from users sketch with geometric sampling. Our generative model is based on Conditional Generative Adversarial Network with the curvature-sensitive points sampling of conditional masks. This sampling process can help eliminate the ambiguity of generation results as network input. In addition, we adopted a U-Net structure discriminator to help the generator be better trained. It is verified that the proposed framework can generate more accurate normal maps., Comment: accepted in International Workshop on Advanced Image Technology 2021, 5 pages, 2 figures

Machine learning potential enables molecular dynamics simulations of systems beyond the capability of classical force fields. The traditional approach to develop structural sets for training machine learning potential typically generate a great number of redundant configurations, which will result in unnecessary computational costs. This work investigates the possibility of reducing redundancy in an initial data set containing 6183 configurations for a Li-Si machine learning potential. Starting from the initial data set, we constructed a series of subsets ranging from 25 to 1500 configurations by combining a structural similarity analysis algorithm and the farthest point sampling method. Results show that the machine learning potential trained from a data set containing 400 configurations can achieve an accuracy comparable to the one developed from the initial data set of 6183 configurations in describing potential energies, atomic forces, and structural properties of Li-Si systems. In addition, the redundancy reducing approach also demonstrates advantages over the classic stochastic method for constructing a concise training data set for Li-Si systems., Comment: 38 pages, 10 figures

Quasiparticle poisoning, expected to arise during the measurement of Majorana zero mode state, poses a fundamental problem towards the realization of Majorana-based quantum computation. Parafermions, a natural generalization of Majorana fermions, can encode topological qudits immune to quasiparticle poisoning. While parafermions are expected to emerge in superconducting fractional quantum Hall systems, they are not yet attainable with current technology. To bypass this problem, we employ a photonic quantum simulator to experimentally demonstrate the key components of parafermion-based universal quantum computation. Our contributions in this article are twofold. First, by manipulating the photonic states, we realize Clifford operator Berry phases that correspond to braiding statistics of parafermions. Second, we investigate the quantum contextuality in a topological system for the first time by demonstrating the contextuality of parafermion encoded qudit states. Importantly, we find that the topologically-encoded contextuality opens the way to magic state distillation, while both the contextuality and the braiding-induced Clifford gates are resilient against local noise. By introducing contextuality, our photonic quantum simulation provides the first step towards a physically robust methodology for realizing topological quantum computation., Comment: 11+6 pages, 5+2 figures, 1+2 tables, presentation extended and improved, analysis and results the same, to appear in PRX Quantum

Recurrent neural transducer (RNN-T) is a promising end-to-end (E2E) model in automatic speech recognition (ASR). It has shown superior performance compared to traditional hybrid ASR systems. However, training RNN-T from scratch is still challenging. We observe a huge gradient variance during RNN-T training and suspect it hurts the performance. In this work, we analyze the cause of the huge gradient variance in RNN-T training and proposed a new \textit{normalized jointer network} to overcome it. We also propose to enhance the RNN-T network with a modified conformer encoder network and transformer-XL predictor networks to achieve the best performance. Experiments are conducted on the open 170-hour AISHELL-1 and industrial-level 30000-hour mandarin speech dataset. On the AISHELL-1 dataset, our RNN-T system gets state-of-the-art results on AISHELL-1's streaming and non-streaming benchmark with CER 6.15\% and 5.37\% respectively. We further compare our RNN-T system with our well trained commercial hybrid system on 30000-hour-industry audio data and get 9\% relative improvement without pre-training or external language model.

In this work we propose an inference technique, asynchronous revision, to unify streaming and non-streaming speech recognition models. Specifically, we achieve dynamic latency with only one model by using arbitrary right context during inference. The model is composed of a stack of convolutional layers for audio encoding. In inference stage, the history states of encoder and decoder can be asynchronously revised to trade off between the latency and the accuracy of the model. To alleviate training and inference mismatch, we propose a training technique, segment cropping, which randomly splits input utterances into several segments with forward connections. This allows us to have dynamic latency speech recognition results with large improvements in accuracy. Experiments show that our dynamic latency model with asynchronous revision gives 8\%-14\% relative improvements over the streaming models.

We present the construction of the $\lambda$-deformation of $AdS_5\times S^5$ superstring from the four dimensional Chern-Simons-type gauge theory. The procedure is applicable to all the semi-symmetric coset models and generalizes the previous construction of $\lambda$-deformation of the symmetric coset models., Comment: match the published version

Fluorescein angiography can provide a map of retinal vascular structure and function, which is commonly used in ophthalmology diagnosis, however, this imaging modality may pose risks of harm to the patients. To help physicians reduce the potential risks of diagnosis, an image translation method is adopted. In this work, we proposed a conditional generative adversarial network(GAN) - based method to directly learn the mapping relationship between structure fundus images and fundus fluorescence angiography images. Moreover, local saliency maps, which define each pixel's importance, are used to define a novel saliency loss in the GAN cost function. This facilitates more accurate learning of small-vessel and fluorescein leakage features., Comment: 16 pages, 6 figures, accepted by Medical Imaging on Deep Learning

Electron-phonon and exciton-phonon interactions in nanoclusters are formulated and computed under the framework of GW-BSE (Bethe-Salpeter equation) approach. The phonon effect is modeled with the two-particle representation for the first time. The nonradiative relaxation rates of electrons and excitons are calculated. It is uncovered that both single-phonon relaxation and multiple-phonon relaxation are significant in nanocrystals, and correspond to two types of physical processes that have totally different spectral lineshapes. Furthermore, the multiple-phonon relaxation always occurs and its rates are comparable to the corresponding single-phonon relaxation rates for both electrons and excitons in the system studied (Si46). The inelastic scattering rates of electrons and excitons are also calculated based on many-body Green function theory. For the electronic states in Si46, the inelastic scattering decay is predicted to be a primary decay mechanism for multiexciton relaxation, and nonradiative relaxation rates are larger than inelastic scattering rates for most excitonic states in Si46., Comment: 43 pages, 14 figures