We devise an articulatory representation-based text-to-speech (TTS) model, ArtSpeech, an explainable and effective network for human-like speech synthesis, by revisiting the sound production system. Current deep TTS models learn acoustic-text mapping in a fully parametric manner, ignoring the explicit physical significance of articulation movement. ArtSpeech, on the contrary, leverages articulatory representations to perform adaptive TTS, clearly describing the voice tone and speaking prosody of different speakers. Specifically, energy, F0, and vocal tract variables are utilized to represent airflow forced by articulatory organs, the degree of tension in the vocal folds of the larynx, and the coordinated movements between different organs, respectively. We also design a multi-dimensional style mapping network to extract speaking styles from the articulatory representations, guided by which variation predictors could predict the final mel spectrogram output. To validate the effectiveness of our approach, we conducted comprehensive experiments and analyses using the widely recognized speech corpus, such as LJSpeech and LibriTTS datasets, yielding promising similarity enhancement between the generated results and the target speaker's voice and prosody.
论文检索
输入标题、作者或关键词，从 4,294 篇学术成果中精准定位
ClickDiff: Click to Induce Semantic Contact Map for Controllable Grasp Generation with Diffusion Models
PDF ↗Grasp generation aims to create complex hand-object interactions with a specified object. While traditional approaches for hand generation have primarily focused on visibility and diversity under scene constraints, they tend to overlook the fine-grained hand-object interactions such as contacts, resulting in inaccurate and undesired grasps. To address these challenges, we propose a controllable grasp generation task and introduce ClickDiff, a controllable conditional generation model that leverages a fine-grained Semantic Contact Map (SCM). Particularly when synthesizing interactive grasps, the method enables the precise control of grasp synthesis through either user-specified or algorithmically predicted Semantic Contact Map. Specifically, to optimally utilize contact supervision constraints and to accurately model the complex physical structure of hands, we propose a Dual Generation Framework. Within this framework, the Semantic Conditional Module generates reasonable contact maps based on fine-grained contact information, while the Contact Conditional Module utilizes contact maps alongside object point clouds to generate realistic grasps. We evaluate the evaluation criteria applicable to controllable grasp generation. Both unimanual and bimanual generation experiments on GRAB and ARCTIC datasets verify the validity of our proposed method, demonstrating the efficacy and robustness of ClickDiff, even with previously unseen objects. Our code is available at https://github.com/adventurer-w/ClickDiff.
Icons are ubiquitous visual elements in graphic design, yet their creation is often complex and time-consuming. To resolve this problem, we draw inspiration from the booming text-to-image field and propose Text-Guided Icon Set Expansion, a novel task that helps users design high-quality icons using textual descriptions. Besides, users can control the style consistency of the created icons by inputting a few hand-crafted icons as style reference. Despite its practicality, the task poses two unique challenges. (i) Abstract Concept Visualization. Abstract concepts like technology and health are frequently encountered in icon creation, but their visualization is not straightforward and requires a grounding process that translates them into physical, easy-to-depict objects. (ii) Fine-grained Style Transfer. Unlike ordinary images, icons exhibit richer fine-grained stylistic elements, including tones, line widths, shapes, shadow effects, etc., which puts higher demands on capturing and preserving detailed styles during icon generation. To address the challenges, we propose IconDM, a method based on pre-trained text-to-image (T2I) diffusion models. Our approach incorporates a one-time domain adaptation process and an online style transfer process. In domain adaptation, we enhance the existing T2I model's capability to understand abstract concepts by fine-tuning it on high-quality icon-text pairs. To achieve this, we construct a large-scale dataset IconBank containing 2.3 million icon samples, and leverage a state-of-the-art vision-language model to generate textual descriptions for each icon. In style transfer, we introduce a Style Enhancement Module into the T2I model. It explicitly extracts the fine-grained style features from the given reference icons and is jointly optimized with the T2I model during DreamBooth tuning. To assess IconDM, we present IconBench, a structured evaluation suite with 30 icon sets and 100 concepts (including 50 abstract concepts). Quantitative results, qualitative analysis, and extensive ablation studies demonstrate the effectiveness of IconDM.
Previous research has demonstrated the potential of Augmented Reality in enhancing psychological comfort in Human-Robot Interaction (AR-HRI) through shared robot intent, enhanced visual feedback, and increased expressiveness and creativity in interaction methods. However, the challenge of selecting interaction methods that enhance physical comfort in varying scenarios remains. This study purposes a dynamic dual-layer interaction adjustment mechanism to improve user comfort and interaction efficiency. The mechanism comprises two models: an general layer model, grounded in ergonomics principles, identifies appropriate areas for various interaction methods; a individual layer model predicts user discomfort levels using physiological signals. Interaction methods are dynamically adjusted based on discomfort level changes, enabling the system to adapt to individual differences and dynamic changes, thereby reducing misjudgments and enhancing comfort management. The mechanism's success in authoring tasks validates its effectiveness, significantly advancing AR-HRI and fostering more comfortable and enhancing efficient human-centered interactions.
E-commerce has emerged as a significant endeavour in which technological advancements influence the shopping experience. Simultaneously, the metaverse is the next breakthrough to transform multimedia engagement. However, under such situations, deceiving designs aimed at deceiving users into making desired choices might be more successful. This paper proposes the design space of manipulative techniques in e-commerce applications for the metaverse. We construct our arguments by evaluating user interaction with manipulative design in metaverse shopping experiences, followed by a survey among users to understand the effect of counteracting manipulative e-commerce scenarios. Our findings can understanding of design guidelines according to metaverse e-commerce experiences and the possibility of opportunities to improve user awareness of manipulative experiences.
As an indispensable personalized service within Location-Based Social Networks (LBSNs), the Point-of-Interest (POI) recommendation aims to assist individuals in discovering attractive and engaging places. However, the accurate recommendation capability relies on the powerful server collecting a vast amount of users' historical check-in data, posing significant risks of privacy breaches. Although several collaborative learning (CL) frameworks for POI recommendation enhance recommendation resilience and allow users to keep personal data on-device, they still share personal knowledge to improve recommendation performance, thus leaving vulnerabilities for potential attackers. Given this, we design a new Physical Trajectory Inference Attack (PTIA) to expose users' historical trajectories. Specifically, for each user, we identify the set of interacted POIs by analyzing the aggregated information from the target POIs and their correlated POIs. We evaluate the effectiveness of PTIA on two real-world datasets across two types of decentralized CL frameworks for POI recommendation. Empirical results demonstrate that PTIA poses a significant threat to users' historical trajectories. Furthermore, Local Differential Privacy (LDP), the traditional privacy-preserving method for CL frameworks, has also been proven ineffective against PTIA. In light of this, we propose a novel defense mechanism (AGD) against PTIA based on an adversarial game to eliminate sensitive POIs and their information in correlated POIs. After conducting intensive experiments, AGD has been proven precise and practical, with minimal impact on recommendation performance.
As a safety-critical application, Autonomous Driving (AD) has received growing attention from security researchers. AD heavily relies on sensors for perception. However, sensors themselves are susceptible to various threats since they are exposed to the environments and vulnerable to malicious or interfering signals. To cope with situations where a sensor might malfunction, Multi Sensor Fusion (MSF) was proposed as a general strategy to enhance the robustness of perception models. In this paper, we focus on investigating MSF security under various sensor attacks and wish to answer the following research questions: (1)Does fusion enhance robustness or not? (2)How does the architecture of the fusion model influence robustness? To this end, we establish a rigorous benchmark for fusion-based 3D object detection robustness. Our new benchmark features 5 types of LiDAR attacks and 6 types of camera attacks. Different from traditional benchmarks, we take the physical sensor attacks into consideration during the corruption construction. Then, we systematically investigate 7 MSF-based and 5 single-modality 3D object detection models with different fusion architectures. We release the benchmarks and codes to facilitate future studies: \textcolorblue \hrefhttps://github.com/Jinzizhisir/PSA-Fusion https://github.com/Jinzizhisir/PSA-Fusion .
Cyber-physical system sensors emit multivariate time series (MTS) that monitor physical system processes. Such time series generally capture unknown numbers of states, each with a different duration, that correspond to specific conditions, e.g., "walking" or "running" in human-activity monitoring. Unsupervised identification of such states facilitates storage and processing in subsequent data analyses, as well as enhances result interpretability. Existing state-detection proposals face three challenges. First, they introduce substantial computational overhead, rendering them impractical in resourceconstrained or streaming settings. Second, although state-of-the-art (SOTA) proposals employ contrastive learning for representation, insufficient attention to false negatives hampers model convergence and accuracy. Third, SOTA proposals predominantly only emphasize offline non-streaming deployment, we highlight an urgent need to optimize online streaming scenarios. We propose E2Usd that enables efficient-yet-accurate unsupervised MTS state detection. E2Usd exploits a Fast Fourier Transform-based Time Series Compressor (fftCompress) and a Decomposed Dual-view Embedding Module (ddEM) that together encode input MTSs at low computational overhead. Additionally, we propose a False Negative Cancellation Contrastive Learning method (fnccLearning) to counteract the effects of false negatives and to achieve more cluster-friendly embedding spaces. To reduce computational overhead further in streaming settings, we introduce Adaptive Threshold Detection (adaTD). Comprehensive experiments with six baselines and six datasets offer evidence that E2Usd is capable of SOTA accuracy at significantly reduced computational overhead. Our code is available at https://github.com/AI4CTS/E2Usd.
Deep learning has brought about a revolutionary transformation in network applications, particularly in domains like e-commerce and online advertising. Distributed training (DT), as a critical means to expedite model training, has progressively emerged as a key foundational infrastructure for such applications. However, with the rapid advancement of hardware accelerators, the performance bottleneck in DT has shifted from computation to communication. In-network aggregation (INA) solutions have shown promise in alleviating the communication bottleneck. Regrettably, current INA solutions primarily focus on improving efficiency under the traditional parameter server (PS) architecture and do not fully address the communication bottleneck caused by limited PS ingress bandwidth. To bridge this gap, we propose InArt, the first work to introduce INA with routing selection in a multi-PS architecture. InArt employs a multi-PS architecture to split DT tasks among multiple PSs, and selects appropriate routing schemes to fully harness INA capabilities. To accommodate traffic dynamics, InArt adopts a two-phase approach: splitting the training model among multiple parameter servers and selecting routing paths for INA. We propose Lagrange multiplier and randomized rounding algorithms for these phases, respectively. We implement InArt and evaluate its performance through experiments on physical platforms (Tofino switches) and Mininet emulation (P4 Software Switches). Experimental results show that InArt can reduce communication time by 48%\!\sim57\!% compared with state-of-the-art solutions.
This paper presents FreqMAE, a novel self-supervised learning framework that synergizes masked autoencoding (MAE) with physics-informed insights to capture feature patterns in multi-modal IoT sensor data. FreqMAE enhances latent space representation of sensor data, reducing reliance on data labeling and improving accuracy for AI tasks. Differing from data augmentation-based methods like contrastive learning, FreqMAE's approach eliminates the need for handcrafted transformations. Adapting MAE for IoT sensing signals, we present three contributions from frequency domain insights: First, a Temporal-Shifting Transformer (TS-T) encoder that enables temporal interactions while distinguishing different frequency bands; Second, a factorized multi-modal fusion mechanism for leveraging cross-modal correlations and preserving unique modality features; Third, a hierarchically weighted loss function that emphasizes important frequency components and high Signal-to-Noise Ratio (SNR) samples. Comprehensive evaluations on two sensing applications validate FreqMAE's proficiency in reducing labeling needs and enhancing resilience against domain shifts.
The European Data Science Day offers a full day focused exclusively on innovative KDD-relevant research and development projects from national and regional funding programs, as well as corporate, start-up, and nonprofit channels. The idea is to bring together a diverse community of researchers in Data Science, Machine Learning, Language Technologies, and Knowledge Discovery, as well as partnerships in the social and physical sciences/arts, to showcase the state-of-the-art in research and applications.
Symbolic regression is a machine learning technique employed for learning mathematical equations directly from data. Mathematical equations capture both functional and causal relationships in the data. In addition, they are simple, compact, generalizable, and interpretable models, making them the best candidates for i) learning inherently transparent models and ii) boosting scientific discovery. Symbolic regression has received a growing interest since the last decade and is tackled using different approaches in supervised and unsupervised deep learning, thanks to the enormous progress achieved in deep learning in the last twenty years. Symbolic regression remains underestimated in conference coverage as a primary form of interpretable AI and a potential candidate for automating scientific discovery. This tutorial overviews symbolic regression: problem definition, approaches, and key limitations, discusses why physical sciences are beneficial to symbolic regression, and explores possible future directions in this research area.
Indoor temperature prediction is crucial for decision-making in central heating systems. Beyond accuracy, predictions shall be interpretable, i.e. conform to the laws of physics; otherwise, it may lead to system failures or unsafe conditions. However, deep learning models often face criticism regarding interpretability, which limits their application in such settings. To this end, we propose a Spatio-Temporal Consistency enhanced Differential Network (CONST) for interpretable indoor temperature prediction. Our approach mainly consists of a differential predictive module and a spatio-temporal consistency module. Modeling the influential factors, the first module solves the issue of multicollinearity through the differential operation. Considering the heterogeneity of global and local data distributions, the second module characterizes the temporal and spatial consistency to mine the universal pattern by multi-task learning, thereby improving the prediction interpretability. Besides, we propose a set of interpretability metrics to overcome the drawbacks of partial dependence plot metric, which are more practical, zero-centered, flexible, and numerical. We conclude experiments on a real-world dataset with four heating stations. The results demonstrate the advantages of our approach over various baselines, where the interpretability can be improved by more than 8 times on cRPD while maintaining high accuracy. We developed CONST on the SmartHeat system, providing hourly indoor temperature forecasts for 13 heating stations in northern China.
Product assortment selection is a critical challenge facing physical retailers. Effectively aligning inventory with the preferences of shoppers can increase sales and decrease out-of-stocks. However, in real-world settings the problem is challenging due to the combinatorial explosion of product assortment possibilities. Consumer preferences are typically heterogeneous across space and time, making inventory-preference alignment challenging. Additionally, existing strategies rely on syndicated data, which tends to be aggregated, low resolution, and suffer from high latency. To solve these challenges, we introduce a real-time recommendation system, which we call EdgeRec3D. Our system utilizes recent advances in 3D computer vision for perception and automatic, fine grained sales estimation. These perceptual components run on the edge of the network and facilitate real-time reward signals. Additionally, we develop a Bayesian payoff model to account for noisy estimates from 3D LIDAR data. We rely on spatial clustering to allow the system to adapt to heterogeneous consumer preferences, and a graph-based candidate generation algorithm to address the combinatorial search problem. We test our system in real-world stores across two, 6-8 week A/B tests with beverage products and demonstrate a 35% and 27% increase in sales respectively. Finally, we monitor the deployed system for a period of 28 weeks with an observational study and show a 9.4% increase in sales.
Incorporating Euclidean symmetries (e.g. rotation equivariance) as inductive biases into graph neural networks has improved their generalization ability and data efficiency in unbounded physical dynamics modeling. However, in various scientific and engineering applications, the symmetries of dynamics are frequently discrete due to the boundary conditions. Thus, existing GNNs either over-look necessary symmetry, resulting in suboptimal representation ability, or impose excessive equivariance, which fails to generalize to unobserved symmetric dynamics. In this work, we propose a general Discrete Equivariant Graph Neural Network (DEGNN) that guarantees equivariance to a given discrete point group. Specifically, we show that such discrete equivariant message passing could be constructed by transforming geometric features into permutation-invariant embeddings. Through relaxing continuous equivariant constraints, DEGNN can employ more geometric feature combinations to approximate unobserved physical object interaction functions. Two implementation approaches of DEGNN are proposed based on ranking or pooling permutation-invariant functions. We apply DEGNN to various physical dynamics, ranging from particle, molecular, crowd to vehicle dynamics. In twenty scenarios, DEGNN significantly outperforms existing state-of-the-art approaches. Moreover, we show that DEGNN is data efficient, learning with less data, and can generalize across scenarios such as unobserved orientation.
Maritime traffic management increasingly relies on vessel position information provided by terrestrial and satellite networks of the Automatic Identification System (AIS). Unfortunately, the problem of missing AIS data can lead to long-term gaps in vessel trajectory, raising corresponding security concerns regarding collision risks and illicit activities. Existing imputation approaches are often constrained by vehicle-based low-sampling trajectories, hindering their ability to address unique characteristics of maritime transportation systems and long-term missing scenarios. To tackle these challenges, we propose a novel generative framework for long-term vessel trajectory imputation. Our framework considers irregular tracks of vessels, which differ from those of cars due to the absence of a structured road network, and ensures the continuity of multi-point imputed trajectories. Specifically, we first utilize a pre-trained trajectory embedding block to capture patterns of vessel movements. Subsequently, we introduce a diffusion-based model for generating missing trajectories, where observed trajectory modeling with transformer encoding architecture and embeddings of both historical vessel trajectory and external factors serve as conditional information. In particular, we design a physics-guided discriminator in the training stage, which imposes kinematic constraints between locations and angles to improve the continuity of the imputed trajectories. Comprehensive experiments and analysis on a real-world AIS dataset confirm the effectiveness of our proposed approach.
Consistent bi-directional inferences are the key for many machine learning applications. Without consistency, inverse learning-based inferences can cause fuzzy images, erroneous control signals, and cascading failure in SCADA systems. Since standard deep neural networks (DNNs) are not inherently invertible to offer consistency, some past methods reconstruct DNN architecture analytically for one-to-one correspondence but compromise key features such as universal approximation. Other work maintains the capability of universal approximation in DNNs via iterative numerical approximation. However, these methods limit their applications significantly due to Lipschitz conditions and issues of numerical convergence. The dilemma of the analytical and numerical methods is the incompatibility between nonlinear layer compositions and bijective function construction for inverse modeling. Based on the observation, we propose decomposed-invertible-pathway DNNs (DipDNN). It relaxes the redundant reconstruction of nested DNN in the former methods and eases the Lipschitz constraint. As a result, we strictly guarantee the consistency of global inverse modeling without harming DNN's capability for universal approximation. As numerical stability and generalizability are keys for controlling critical infrastructures, we integrate contractive property with a parallel structure for inductive biases, leading to stable performance. Numerical results show that DipDNN performs significantly better than past methods, thanks to its enforcement of inverse consistency, numerical stability, and physical regularization.
In this paper, we study cascading failures in power grids through the lens of information diffusion models. Similar to the spread of rumors or influence in an online social network, it has been observed that failures (outages) in a power grid can spread contagiously, driven by viral spread mechanisms. We employ a stochastic diffusion model that is Markovian (memoryless) and local (the activation of one node, i.e., transmission line, can only be caused by its neighbors). Our model integrates viral diffusion principles with physics-based concepts, by correlating the diffusion weights (contagion probabilities between transmission lines) with the hyperparametric Information Cascades (IC) model. We show that this diffusion model can be learned from traces of cascading failures, enabling accurate modeling and prediction of failure propagation. This approach facilitates actionable information through well-understood and efficient graph analysis methods and graph diffusion simulations. Furthermore, by leveraging the hyperparametric model, we can predict diffusion and mitigate the risks of cascading failures even in unseen grid configurations, whereas existing methods falter due to a lack of training data. Extensive experiments based on a benchmark power grid and simulations therein show that our approach effectively captures the failure diffusion phenomena and guides decisions to strengthen the grid, reducing the risk of large-scale cascading failures. Additionally, we characterize our model's sample complexity, improving upon the existing bound.
Modeling the evolution of physical dynamics is a foundational problem in science and engineering, and it is regarded as the modeling of an operator mapping between infinite-dimensional functional spaces. Operator learning methods, learning the underlying infinite-dimensional operator in a high-dimensional latent space, have shown significant potential in modeling physical dynamics. However, there remains insufficient research on how to approximate an infinite-dimensional operator using a finite-dimensional parameter space. Inappropriate dimensionality representation of the underlying operator leads to convergence difficulties, decreasing generalization capability, and violating the physical consistency. To address the problem, we present Neural Manifold Operator (NMO) to learn the invariant subspace with the intrinsic dimension to parameterize infinite-dimensional underlying operators. NMO achieves state-of-the-art performance in statistical and physical metrics and gains 23.35% average improvement on three real-world scenarios and four equation-governed scenarios across a wide range of multi-disciplinary fields. Our paradigm has demonstrated universal effectiveness across various model structure implementations, including Multi-Layer Perceptron, Convolutional Neural Networks, and Transformers. Experimentally, we prove that the intrinsic dimension calculated by our paradigm is the optimal dimensional representation of the underlying operators. We release our code at https://github.com/AI4EarthLab/Neural-Manifold-Operators.
Learning from Emergence: A Study on Proactively Inhibiting the Monosemantic Neurons of Artificial Neural Networks
PDF ↗Recently, emergence has received widespread attention from the research community along with the success of large-scale models. Different from the literature, we hypothesize a key factor that promotes the performance during the increase of scale: the reduction of monosemantic neurons that can only form one-to-one correlations with specific features. Monosemantic neurons tend to be sparser and have negative impacts on the performance in large models. Inspired by this insight, we propose an intuitive idea to identify monosemantic neurons and inhibit them. However, achieving this goal is a non-trivial task as there is no unified quantitative evaluation metric and simply banning monosemantic neurons does not promote polysemanticity in neural networks. Therefore, we first propose a new metric to measure the monosemanticity of neurons with the guarantee of efficiency for online computation, then introduce a theoretically supported method to suppress monosemantic neurons and proactively promote the ratios of polysemantic neurons in training neural networks. We validate our conjecture that monosemanticity brings about performance change at different model scales on a variety of neural networks and benchmark datasets in different areas, including language, image, and physics simulation tasks. Further experiments validate our analysis and theory regarding the inhibition of monosemanticity.