Search

The text-to-video (T2V) generation models, offering convenient visual creation, have recently garnered increasing attention. Despite their substantial potential, the generated videos may present artifacts, including structural implausibility, temporal inconsistency, and a lack of motion, often resulting in near-static video. In this work, we have identified a correlation between the disparity of temporal attention maps across different blocks and the occurrence of temporal inconsistencies. Additionally, we have observed that the energy contained within the temporal attention maps is directly related to the magnitude of motion amplitude in the generated videos. Based on these observations, we present BroadWay, a training-free method to improve the quality of text-to-video generation without introducing additional parameters, augmenting memory or sampling time. Specifically, BroadWay is composed of two principal components: 1) Temporal Self-Guidance improves the structural plausibility and temporal consistency of generated videos by reducing the disparity between the temporal attention maps across various decoder blocks. 2) Fourier-based Motion Enhancement enhances the magnitude and richness of motion by amplifying the energy of the map. Extensive experiments demonstrate that BroadWay significantly improves the quality of text-to-video generation with negligible additional cost.

High-speed ultra-broadband detectors play a crucial role in aerospace technology, and national security etc. The interfacial work function internal photoemission (IWIP) detector employs multiple absorption mechanism comprehensively across different wavelength band to achieve complete photon type detection, which makes it possible to realize high-speed and ultra-broadband simultaneously. We propose a ratchet heterojunction IWIP (HEIWIP) detector, which shows 3-165THz ultra-broadband coverage. The high-speed response is investigated in detail by both microwave rectification technology and high-speed modulated terahertz light. Up to 5.1GHz 3dB bandwidth is acquired in terms of microwave rectification measurement. And 4.255GHz inter-mode optical beat note signal was successfully detected.

Retrieval-Augmented Generation (RAG) is widely used to inject external non-parametric knowledge into large language models (LLMs). Recent works suggest that Knowledge Graphs (KGs) contain valuable external knowledge for LLMs. Retrieving information from KGs differs from extracting it from document sets. Most existing approaches seek to directly retrieve relevant subgraphs, thereby eliminating the need for extensive SPARQL annotations, traditionally required by semantic parsing methods. In this paper, we model the subgraph retrieval task as a conditional generation task handled by small language models. Specifically, we define a subgraph identifier as a sequence of relations, each represented as a special token stored in the language models. Our base generative subgraph retrieval model, consisting of only 220M parameters, achieves competitive retrieval performance compared to state-of-the-art models relying on 7B parameters, demonstrating that small language models are capable of performing the subgraph retrieval task. Furthermore, our largest 3B model, when plugged with an LLM reader, sets new SOTA end-to-end performance on both the WebQSP and CWQ benchmarks. Our model and data will be made available online: https://github.com/hwy9855/GSR., Comment: Accepted by EMNLP 2024 Findings

In this article, we present results of our recent work of epitaxially-grown aluminum (epi-Al) on antimonide heterostructures, where the epi-Al thin film is grown at either room temperature or below zero $^o$C. A sharp superconducting transition at $T \sim 1.3$ K is observed in these epi-Al films, and the critical magnetic field follows the BCS (Bardeen-Cooper-Schrieffer) model. We further show that supercurrent states are achieved in Josephson junctions fabricated in the epi-Al/antimonide heterostructures with mobility $\mu \sim 1.0 \times 10^6$ cm$^2$/Vs, making these heterostructures a promising platform for the exploration of Josephson junction effects for quantum microelectronics applications, and the realization of robust topological superconducting states that potentially allow the realization of intrinsically fault-tolerant qubits and quantum gates.

3D Gaussian Splatting is capable of reconstructing 3D scenes in minutes. Despite recent advances in improving surface reconstruction accuracy, the reconstructed results still exhibit bias and suffer from inefficiency in storage and training. This paper provides a different observation on the cause of the inefficiency and the reconstruction bias, which is attributed to the integration of the low-opacity parts (LOPs) of the generated Gaussians. We show that LOPs consist of Gaussians with overall low-opacity (LOGs) and the low-opacity tails (LOTs) of Gaussians. We propose Spiking GS to reduce such two types of LOPs by integrating spiking neurons into the Gaussian Splatting pipeline. Specifically, we introduce global and local full-precision integrate-and-fire spiking neurons to the opacity and representation function of flattened 3D Gaussians, respectively. Furthermore, we enhance the density control strategy with spiking neurons' thresholds and a new criterion on the scale of Gaussians. Our method can represent more accurate reconstructed surfaces at a lower cost. The supplementary material and code are available at https://github.com/zju-bmi-lab/SpikingGS.

Cross-domain Recommendation systems leverage multi-domain user interactions to improve performance, especially in sparse data or new user scenarios. However, CDR faces challenges such as effectively capturing user preferences and avoiding negative transfer. To address these issues, we propose the Multi-view Disentangled and Adaptive Preference Learning (MDAP) framework. Our MDAP framework uses a multiview encoder to capture diverse user preferences. The framework includes a gated decoder that adaptively combines embeddings from different views to generate a comprehensive user representation. By disentangling representations and allowing adaptive feature selection, our model enhances adaptability and effectiveness. Extensive experiments on benchmark datasets demonstrate that our method significantly outperforms state-of-the-art CDR and single-domain models, providing more accurate recommendations and deeper insights into user behavior across different domains., Comment: The International Web Information Systems Engineering conference

As machine learning models continue to swiftly advance, calibrating their performance has become a major concern prior to practical and widespread implementation. Most existing calibration methods often negatively impact model accuracy due to the lack of diversity of validation data, resulting in reduced generalizability. To address this, we propose a calibration method that incorporates synthetic data without compromising accuracy. We derive the expected calibration error (ECE) bound using the Probably Approximately Correct (PAC) learning framework. Large language models (LLMs), known for their ability to mimic real data and generate text with mixed class labels, are utilized as a synthetic data generation strategy to lower the ECE bound and improve model accuracy on real test data. Additionally, we propose data generation mechanisms for efficient calibration. Testing our method on four different natural language processing tasks, we observed an average up to 34\% increase in accuracy and 33\% decrease in ECE., Comment: Accepted to EMNLP 2024 Main Conference (Long paper)

Imaging through thick scattering media presents significant challenges, particularly for three-dimensional (3D) applications. This manuscript demonstrates a novel scheme for single-image-enabled 3D imaging through such media, treating the scattering medium as a lens. This approach captures a comprehensive image containing information about objects hidden at various depths. By leveraging depth from focus and the reduced thickness of the scattering medium for single-pixel imaging, the proposed method ensures robust 3D imaging capabilities. We develop both traditional metric-based and deep learning-based methods to extract depth information for each pixel, allowing us to explore the locations of both positive and negative objects, whether shallow or deep. Remarkably, this scheme enables the simultaneous 3D reconstruction of targets concealed within the scattering medium. Specifically, we successfully reconstructed targets buried at depths of 5 mm and 30 mm within a total medium thickness of 60 mm. Additionally, we can effectively distinguish targets at three different depths. Notably, this scheme requires no prior knowledge of the scattering medium, no invasive procedures, reference measurements, or calibration.

Integrated energy systems (IESs) are complex systems consisting of diverse operating units spanning multiple domains. To address its operational challenges, we propose a physics-informed hybrid time-series neural network (NN) surrogate to predict the dynamic performance of IESs across multiple time scales. This neural network-based modeling approach develops time-series multi-layer perceptrons (MLPs) for the operating units and integrates them with prior process knowledge about system structure and fundamental dynamics. This integration forms three hybrid NNs (long-term, slow, and fast MLPs) that predict the entire system dynamics across multiple time scales. Leveraging these MLPs, we design an NN-based scheduler and an NN-based economic model predictive control (NEMPC) framework to meet global operational requirements: rapid electrical power responsiveness to operators requests, adequate cooling supply to customers, and increased system profitability, while addressing the dynamic time-scale multiplicity present in IESs. The proposed day-ahead scheduler is formulated using the ReLU network-based MLP, which effectively represents IES performance under a broad range of conditions from a long-term perspective. The scheduler is then exactly recast into a mixed-integer linear programming problem for efficient evaluation. The real-time NEMPC, based on slow and fast MLPs, comprises two sequential distributed control agents: a slow NEMPC for the cooling-dominant subsystem with slower transient responses and a fast NEMPC for the power-dominant subsystem with faster responses. Extensive simulations demonstrate that the developed scheduler and NEMPC schemes outperform their respective benchmark scheduler and controller by about 25% and 40%. Together, they enhance overall system performance by over 70% compared to benchmark approaches.

Spatiotemporal neural networks have shown great promise in urban scenarios by effectively capturing temporal and spatial correlations. However, urban environments are constantly evolving, and current model evaluations are often limited to traffic scenarios and use data mainly collected only a few weeks after training period to evaluate model performance. The generalization ability of these models remains largely unexplored. To address this, we propose a Spatiotemporal Out-of-Distribution (ST-OOD) benchmark, which comprises six urban scenario: bike-sharing, 311 services, pedestrian counts, traffic speed, traffic flow, ride-hailing demand, and bike-sharing, each with in-distribution (same year) and out-of-distribution (next years) settings. We extensively evaluate state-of-the-art spatiotemporal models and find that their performance degrades significantly in out-of-distribution settings, with most models performing even worse than a simple Multi-Layer Perceptron (MLP). Our findings suggest that current leading methods tend to over-rely on parameters to overfit training data, which may lead to good performance on in-distribution data but often results in poor generalization. We also investigated whether dropout could mitigate the negative effects of overfitting. Our results showed that a slight dropout rate could significantly improve generalization performance on most datasets, with minimal impact on in-distribution performance. However, balancing in-distribution and out-of-distribution performance remains a challenging problem. We hope that the proposed benchmark will encourage further research on this critical issue.

Combining Atacama Large Millimeter/sub-millimeter Array CO(2-1) mapping and JWST near- and mid-infrared imaging, we characterize the relationship between CO(2-1) and polycyclic aromatic hydrocarbon (PAH) emission at ~100 pc resolution in 66 nearby star-forming galaxies, expanding the sample size from previous ~100 pc resolution studies by more than an order of magnitude. Focusing on regions of galaxies where most of the gas is likely to be molecular, we find strong correlations between CO(2-1) and 3.3 micron, 7.7 micron, and 11.3 micron PAH emission, estimated from JWST's F335M, F770W, and F1130W filters. We derive power law relations between CO(2-1) and PAH emission, which have indices in the range 0.8-1.2, implying relatively weak variations in the observed CO-to-PAH ratios across the regions that we study. We find that CO-to-PAH ratios and scaling relationships near HII regions are similar to those in diffuse sight lines. The main difference between the two types of regions is that sight lines near HII regions show higher intensities in all tracers. Galaxy centers, on the other hand, show higher overall intensities and enhanced CO-to-PAH ratios compared to galaxy disks. Individual galaxies show 0.19 dex scatter in the normalization of CO at fixed I_PAH and this normalization anti-correlates with specific star formation rate (SFR/M*) and correlates with stellar mass. We provide a prescription that accounts for these galaxy-to-galaxy variations and represents our best current empirical predictor to estimate CO(2-1) intensity from PAH emission, which allows one to take advantage of JWST's excellent sensitivity and resolution to trace cold gas., Comment: 21 pages, 4 figures, 3 tables. Submitted to ApJ

Diffusion-based 2D virtual try-on (VTON) techniques have recently demonstrated strong performance, while the development of 3D VTON has largely lagged behind. Despite recent advances in text-guided 3D scene editing, integrating 2D VTON into these pipelines to achieve vivid 3D VTON remains challenging. The reasons are twofold. First, text prompts cannot provide sufficient details in describing clothing. Second, 2D VTON results generated from different viewpoints of the same 3D scene lack coherence and spatial relationships, hence frequently leading to appearance inconsistencies and geometric distortions. To resolve these problems, we introduce an image-prompted 3D VTON method (dubbed GS-VTON) which, by leveraging 3D Gaussian Splatting (3DGS) as the 3D representation, enables the transfer of pre-trained knowledge from 2D VTON models to 3D while improving cross-view consistency. (1) Specifically, we propose a personalized diffusion model that utilizes low-rank adaptation (LoRA) fine-tuning to incorporate personalized information into pre-trained 2D VTON models. To achieve effective LoRA training, we introduce a reference-driven image editing approach that enables the simultaneous editing of multi-view images while ensuring consistency. (2) Furthermore, we propose a persona-aware 3DGS editing framework to facilitate effective editing while maintaining consistent cross-view appearance and high-quality 3D geometry. (3) Additionally, we have established a new 3D VTON benchmark, 3D-VTONBench, which facilitates comprehensive qualitative and quantitative 3D VTON evaluations. Through extensive experiments and comparative analyses with existing methods, the proposed \OM has demonstrated superior fidelity and advanced editing capabilities, affirming its effectiveness for 3D VTON., Comment: 21 pages, 11 figures

Reinforcement learning from human feedback (RLHF) methods are emerging as a way to fine-tune diffusion models (DMs) for visual generation. However, commonly used on-policy strategies are limited by the generalization capability of the reward model, while off-policy approaches require large amounts of difficult-to-obtain paired human-annotated data, particularly in visual generation tasks. To address the limitations of both on- and off-policy RLHF, we propose a preference optimization method that aligns DMs with preferences without relying on reward models or paired human-annotated data. Specifically, we introduce a Semi-Policy Preference Optimization (SePPO) method. SePPO leverages previous checkpoints as reference models while using them to generate on-policy reference samples, which replace "losing images" in preference pairs. This approach allows us to optimize using only off-policy "winning images." Furthermore, we design a strategy for reference model selection that expands the exploration in the policy space. Notably, we do not simply treat reference samples as negative examples for learning. Instead, we design an anchor-based criterion to assess whether the reference samples are likely to be winning or losing images, allowing the model to selectively learn from the generated reference samples. This approach mitigates performance degradation caused by the uncertainty in reference sample quality. We validate SePPO across both text-to-image and text-to-video benchmarks. SePPO surpasses all previous approaches on the text-to-image benchmarks and also demonstrates outstanding performance on the text-to-video benchmarks. Code will be released in https://github.com/DwanZhang-AI/SePPO.

DNA-based data storage has been attracting significant attention due to its extremely high density, low power consumption, and long duration compared to traditional data storage mediums. Despite the recent advancements in DNA data storage technology, significant challenges remain. In particular, various types of errors can occur during the processes of DNA synthesis, storage, and sequencing, including substitution errors, insertion errors, and deletion errors. Furthermore, the entire oligo may be lost. In this work, we report a DNA-based data storage architecture that incorporates efficient channel coding schemes, including different types of error-correcting codes (ECCs) and constrained codes, for both the inner coding and outer coding for the DNA data storage channel. We also carry out large scale experiments to validate our proposed DNA data storage architecture. Specifically, 1.61 and 1.69 MB data are encoded into 30,000 oligos each, with information densities of 1.731 and 1.815, respectively. It has been found that the stored information can be fully recovered without any error at average coverages 4.5 and 6.0, respectively. Compared to previous experimental studies, our architecture achieves higher information density and lower coverage, demonstrating the efficiency of the proposed channel coding schemes.

As a physical layer security technology, directional modulation (DM) can be combined with intelligent reflect-ing surface (IRS) to improve the security of drone communications. In this paper, a directional modulation scheme assisted by the IRS is proposed to maximize the transmission rate of unmanned aerial vehicle (UAV) secure communication. Specifically, with the assistance of the IRS, the UAV transmits legitimate information and main-tains its constellation pattern at the location of legitimate users on the ground, while the constellation pattern is disrupted at the eavesdropper's location. In order to solve the joint optimization problem of digital weight coefficients, UAV position, and IRS discrete phase shift, firstly, the digital weight vector and UAV position are optimized through power minimization. Secondly, three methods are proposed to optimize IRS phase shift, namely vector trajectory (VT) method, cross entropy vector trajectory (CE-VT) algorithm, and block coordinate descent vector trajectory (BCD-VT) algorithm. Compared to traditional cross entropy (CE) methods and block coordinate descent (BCD) methods, the proposed CE-VT and BCD-VT algorithms can improve transmission rate performance. The numerical results validate the effectiveness of the optimization scheme in IRS assisted UAV communication.

Text-to-image (T2I) models are increasingly used in impactful real-life applications. As such, there is a growing need to audit these models to ensure that they generate desirable, task-appropriate images. However, systematically inspecting the associations between prompts and generated content in a human-understandable way remains challenging. To address this, we propose \emph{Concept2Concept}, a framework where we characterize conditional distributions of vision language models using interpretable concepts and metrics that can be defined in terms of these concepts. This characterization allows us to use our framework to audit models and prompt-datasets. To demonstrate, we investigate several case studies of conditional distributions of prompts, such as user defined distributions or empirical, real world distributions. Lastly, we implement Concept2Concept as an open-source interactive visualization tool facilitating use by non-technical end-users. Warning: This paper contains discussions of harmful content, including CSAM and NSFW material, which may be disturbing to some readers.

Data attribution methods aim to quantify the influence of individual training samples on the prediction of artificial intelligence (AI) models. As training data plays an increasingly crucial role in the modern development of large-scale AI models, data attribution has found broad applications in improving AI performance and safety. However, despite a surge of new data attribution methods being developed recently, there lacks a comprehensive library that facilitates the development, benchmarking, and deployment of different data attribution methods. In this work, we introduce $\texttt{dattri}$, an open-source data attribution library that addresses the above needs. Specifically, $\texttt{dattri}$ highlights three novel design features. Firstly, $\texttt{dattri}$ proposes a unified and easy-to-use API, allowing users to integrate different data attribution methods into their PyTorch-based machine learning pipeline with a few lines of code changed. Secondly, $\texttt{dattri}$ modularizes low-level utility functions that are commonly used in data attribution methods, such as Hessian-vector product, inverse-Hessian-vector product or random projection, making it easier for researchers to develop new data attribution methods. Thirdly, $\texttt{dattri}$ provides a comprehensive benchmark framework with pre-trained models and ground truth annotations for a variety of benchmark settings, including generative AI settings. We have implemented a variety of state-of-the-art efficient data attribution methods that can be applied to large-scale neural network models, and will continuously update the library in the future. Using the developed $\texttt{dattri}$ library, we are able to perform a comprehensive and fair benchmark analysis across a wide range of data attribution methods. The source code of $\texttt{dattri}$ is available at https://github.com/TRAIS-Lab/dattri.

Gaussian Processes (GPs) are vital for modeling and predicting irregularly-spaced, large geospatial datasets. However, their computations often pose significant challenges in large-scale applications. One popular method to approximate GPs is the Vecchia approximation, which approximates the full likelihood via a series of conditional probabilities. The classical Vecchia approximation uses univariate conditional distributions, which leads to redundant evaluations and memory burdens. To address this challenge, our study introduces block Vecchia, which evaluates each multivariate conditional distribution of a block of observations, with blocks formed using the K-means algorithm. The proposed GPU framework for the block Vecchia uses varying batched linear algebra operations to compute multivariate conditional distributions concurrently, notably diminishing the frequent likelihood evaluations. Diving into the factor affecting the accuracy of the block Vecchia, the neighbor selection criterion is investigated, where we found that the random ordering markedly enhances the approximated quality as the block count becomes large. To verify the scalability and efficiency of the algorithm, we conduct a series of numerical studies and simulations, demonstrating their practical utility and effectiveness compared to the exact GP. Moreover, we tackle large-scale real datasets using the block Vecchia method, i.e., high-resolution 3D profile wind speed with a million points.

Large Language Models (LLMs) have demonstrated remarkable capabilities across various fields, from natural language understanding to text generation. Compared to non-generative LLMs like BERT and DeBERTa, generative LLMs like GPT series and Llama series are currently the main focus due to their superior algorithmic performance. The advancements in generative LLMs are closely intertwined with the development of hardware capabilities. Various hardware platforms exhibit distinct hardware characteristics, which can help improve LLM inference performance. Therefore, this paper comprehensively surveys efficient generative LLM inference on different hardware platforms. First, we provide an overview of the algorithm architecture of mainstream generative LLMs and delve into the inference process. Then, we summarize different optimization methods for different platforms such as CPU, GPU, FPGA, ASIC, and PIM/NDP, and provide inference results for generative LLMs. Furthermore, we perform a qualitative and quantitative comparison of inference performance with batch sizes 1 and 8 on different hardware platforms by considering hardware power consumption, absolute inference speed (tokens/s), and energy efficiency (tokens/J). We compare the performance of the same optimization methods across different hardware platforms, the performance across different hardware platforms, and the performance of different methods on the same hardware platform. This provides a systematic and comprehensive summary of existing inference acceleration work by integrating software optimization methods and hardware platforms, which can point to the future trends and potential developments of generative LLMs and hardware technology for edge-side scenarios., Comment: 43 pages, 15 figures

The rapid advancement of large language models (LLMs) has significantly enhanced the capabilities of AI-driven agents across various tasks. However, existing agentic systems, whether based on fixed pipeline algorithms or pre-defined meta-learning frameworks, cannot search the whole agent design space due to the restriction of human-designed components, and thus might miss the globally optimal agent design. In this paper, we introduce G\"odel Agent, a self-evolving framework inspired by the G\"odel machine, enabling agents to recursively improve themselves without relying on predefined routines or fixed optimization algorithms. G\"odel Agent leverages LLMs to dynamically modify its own logic and behavior, guided solely by high-level objectives through prompting. Experimental results on mathematical reasoning and complex agent tasks demonstrate that implementation of G\"odel Agent can achieve continuous self-improvement, surpassing manually crafted agents in performance, efficiency, and generalizability., Comment: Work in progress

With the rapid advancement of large language models (LLMs), recent years have witnessed many promising studies on leveraging LLM-based agents to simulate human social behavior. While prior work has demonstrated significant potential across various domains, much of it has focused on specific scenarios involving a limited number of agents and has lacked the ability to adapt when errors occur during simulation. To overcome these limitations, we propose a novel LLM-agent-based simulation platform called \textit{GenSim}, which: (1) \textbf{Abstracts a set of general functions} to simplify the simulation of customized social scenarios; (2) \textbf{Supports one hundred thousand agents} to better simulate large-scale populations in real-world contexts; (3) \textbf{Incorporates error-correction mechanisms} to ensure more reliable and long-term simulations. To evaluate our platform, we assess both the efficiency of large-scale agent simulations and the effectiveness of the error-correction mechanisms. To our knowledge, GenSim represents an initial step toward a general, large-scale, and correctable social simulation platform based on LLM agents, promising to further advance the field of social science.

Knowledge distillation aims to transfer knowledge from a large teacher model to a compact student counterpart, often coming with a significant performance gap between them. We find that a too-large performance gap can hamper the training process, which is also verified in recent studies. To address this, we propose a Gap Preserving Distillation (GPD) method that trains an additional dynamic teacher model from scratch along with training the student to bridge this gap. In this way, it becomes possible to maintain a reasonable performance gap between teacher and student during the whole distillation process. To further strengthen distillation from the dynamic teacher to the student, we develop a hard strategy by enforcing them to share parameters and encouraging parameter inheritance. Besides hard strategy, we also build the soft bidirectional mappings between them which are built on an Inverse Reparameterization (IR) method and a Channel-Branch Reparameterization (CBR) strategy. We highlight that our IR is able to initialize a larger dynamic teacher with an arbitrary expansion ratio, while preserving exactly the same accuracy as the given student model. In this way, it guarantees that the dynamic teacher and student start from the same point and avoid a too large gap in early stage of training. As for our CBR, with parameter-sharing, it directly extracts an effective student model from the well-learned dynamic teacher without any post-training, making our method highly flexible for model deployment. In the experiments, GPD significantly outperforms existing distillation methods on top of both CNNs and transformers architectures, achieving up to 1.58% accuracy improvement. Interestingly, GPD also generalizes well to the scenarios without a pre-trained teacher, including training from scratch and fine-tuning, yielding a large improvement of 1.80% and 0.89% on ResNet18, respectively., Comment: 10 pages for the main paper

Understanding how a flow turns into an amorphous solid is a fundamental challenge in statistical physics, during which no apparent structural ordering appears. In the athermal limit, the two states are connected by a well-defined jamming transition, near which the solid is marginally stable. A recent mechanical response screening theory proposes an additional transition above jamming, called a plastic-to-elastic transition here, separating anomalous and quasi-elastic mechanical behavior. Through numerical inflation simulations in two dimensions, we show that the onsets of long-range radial and angular correlations of particle displacements decouple, occurring respectively at the jamming and plastic-to-elastic transitions. The latter is characterized by a power-law diverging correlation angle and a power-law spectrum of the displacements along a circle. This work establishes two-step transitions on the mechanical properties during ``decompression melting'' of an athermal over-jammed amorphous solid, reminiscent of the two-step structural melting of a crystal in two dimensions. In contradistinction with the latter, the plastic-to-elastic transition exists also in three dimensions., Comment: 7 pages, 5 figures (SI 3 pages, 3 figures)

The increasing capability of large language models (LLMs) to generate fluent long-form texts is presenting new challenges in distinguishing machine-generated outputs from human-written ones, which is crucial for ensuring authenticity and trustworthiness of expressions. Existing zero-shot detectors primarily focus on token-level distributions, which are vulnerable to real-world domain shifts, including different prompting and decoding strategies, and adversarial attacks. We propose a more robust method that incorporates abstract elements, such as event transitions, as key deciding factors to detect machine versus human texts by training a latent-space model on sequences of events or topics derived from human-written texts. In three different domains, machine-generated texts, which are originally inseparable from human texts on the token level, can be better distinguished with our latent-space model, leading to a 31% improvement over strong baselines such as DetectGPT. Our analysis further reveals that, unlike humans, modern LLMs like GPT-4 generate event triggers and their transitions differently, an inherent disparity that helps our method to robustly detect machine-generated texts.

Deploying large language models (LLMs) locally on mobile devices is advantageous in scenarios where transmitting data to remote cloud servers is either undesirable due to privacy concerns or impractical due to network connection. Recent advancements (MLC, 2023a; Gerganov, 2023) have facilitated the local deployment of LLMs. However, local deployment also presents challenges, particularly in balancing quality (generative performance), latency, and throughput within the hardware constraints of mobile devices. In this paper, we introduce our lightweight, all-in-one automated benchmarking framework that allows users to evaluate LLMs on mobile devices. We provide a comprehensive benchmark of various popular LLMs with different quantization configurations (both weights and activations) across multiple mobile platforms with varying hardware capabilities. Unlike traditional benchmarks that assess full-scale models on high-end GPU clusters, we focus on evaluating resource efficiency (memory and power consumption) and harmful output for compressed models on mobile devices. Our key observations include i) differences in energy efficiency and throughput across mobile platforms; ii) the impact of quantization on memory usage, GPU execution time, and power consumption; and iii) accuracy and performance degradation of quantized models compared to their non-quantized counterparts; and iv) the frequency of hallucinations and toxic content generated by compressed LLMs on mobile devices., Comment: 10 pages

In recent years, artificial intelligence (AI) has become increasingly integrated into education, reshaping traditional learning environments. Despite this, there has been limited investigation into fully operational artificial human lecturers. To the best of our knowledge, our paper presents the world's first study examining their deployment in a real-world educational setting. Specifically, we investigate the use of "digital teachers," AI-powered virtual lecturers, in a postgraduate course at the Hong Kong University of Science and Technology (HKUST). Our study explores how features such as appearance, non-verbal cues, voice, and verbal expression impact students' learning experiences. Findings suggest that students highly value naturalness, authenticity, and interactivity in digital teachers, highlighting areas for improvement, such as increased responsiveness, personalized avatars, and integration with larger learning platforms. We conclude that digital teachers have significant potential to enhance education by providing a more flexible, engaging, personalized, and accessible learning experience for students., Comment: 28 pages, 6 figures (10 sub-figures), 3 tables

Rate control algorithms are at the heart of video conferencing platforms, determining target bitrates that match dynamic network characteristics for high quality. Recent data-driven strategies have shown promise for this challenging task, but the performance degradation they introduce during training has been a nonstarter for many production services, precluding adoption. This paper aims to bolster the practicality of data-driven rate control by presenting an alternative avenue for experiential learning: leveraging purely existing telemetry logs produced by the incumbent algorithm in production. We observe that these logs contain effective decisions, although often at the wrong times or in the wrong order. To realize this approach despite the inherent uncertainty that log-based learning brings (i.e., lack of feedback for new decisions), our system, Tarzan, combines a variety of robust learning techniques (i.e., conservatively reasoning about alternate behavior to minimize risk and using a richer model formulation to account for environmental noise). Across diverse networks (emulated and real-world), Tarzan outperforms the widely deployed GCC algorithm, increasing average video bitrates by 15-39% while reducing freeze rates by 60-100%.

Ontologies often lack explicit disjointness declarations between classes, despite their usefulness for sophisticated reasoning and consistency checking in Knowledge Graphs. In this study, we explore the potential of Large Language Models (LLMs) to enrich ontologies by identifying and asserting class disjointness axioms. Our approach aims at leveraging the implicit knowledge embedded in LLMs, using prompt engineering to elicit this knowledge for classifying ontological disjointness. We validate our methodology on the DBpedia ontology, focusing on open-source LLMs. Our findings suggest that LLMs, when guided by effective prompt strategies, can reliably identify disjoint class relationships, thus streamlining the process of ontology completion without extensive manual input. For comprehensive disjointness enrichment, we propose a process that takes logical relationships between disjointness and subclass statements into account in order to maintain satisfiability and reduce the number of calls to the LLM. This work provides a foundation for future applications of LLMs in automated ontology enhancement and offers insights into optimizing LLM performance through strategic prompt design. Our code is publicly available on GitHub at https://github.com/n28div/llm-disjointness., Comment: Accepted at KBC-LM'24 workshop at ISWC 2024

We investigate the motion of particles in the spacetime of a Kerr black hole immersed in swirling universes. Using the Poincar\'{e} section, fast Lyapunov exponent indicator, bifurcation diagram and basins of attraction, we present the effects of the swirling parameter and the spin parameter on the dynamical behaviors of the motion of particles, and confirm the presence of chaos in the motion of particles in this background spacetime. We find that the swirling parameter can change the range of the spin parameter where the chaos occurs, and vice versa. Moreover, we observe clearly that, regardless of the spin parameter, there exist some self-similar fractal fine structures in the basins boundaries of attractors for the spacetime of a black hole immersed in swirling universes. The combination the swirling parameter and the spin parameter provides richer physics in the motion of particles., Comment: 17 pages, 12 figures

Text watermarking for Large Language Models (LLMs) has made significant progress in detecting LLM outputs and preventing misuse. Current watermarking techniques offer high detectability, minimal impact on text quality, and robustness to text editing. However, current researches lack investigation into the imperceptibility of watermarking techniques in LLM services. This is crucial as LLM providers may not want to disclose the presence of watermarks in real-world scenarios, as it could reduce user willingness to use the service and make watermarks more vulnerable to attacks. This work is the first to investigate the imperceptibility of watermarked LLMs. We design an identification algorithm called Water-Probe that detects watermarks through well-designed prompts to the LLM. Our key motivation is that current watermarked LLMs expose consistent biases under the same watermark key, resulting in similar differences across prompts under different watermark keys. Experiments show that almost all mainstream watermarking algorithms are easily identified with our well-designed prompts, while Water-Probe demonstrates a minimal false positive rate for non-watermarked LLMs. Finally, we propose that the key to enhancing the imperceptibility of watermarked LLMs is to increase the randomness of watermark key selection. Based on this, we introduce the Water-Bag strategy, which significantly improves watermark imperceptibility by merging multiple watermark keys., Comment: 25 pages, 5 figures, 8 tables

Echocardiography (ECHO) is essential for cardiac assessments, but its video quality and interpretation heavily relies on manual expertise, leading to inconsistent results from clinical and portable devices. ECHO video generation offers a solution by improving automated monitoring through synthetic data and generating high-quality videos from routine health data. However, existing models often face high computational costs, slow inference, and rely on complex conditional prompts that require experts' annotations. To address these challenges, we propose ECHOPULSE, an ECG-conditioned ECHO video generation model. ECHOPULSE introduces two key advancements: (1) it accelerates ECHO video generation by leveraging VQ-VAE tokenization and masked visual token modeling for fast decoding, and (2) it conditions on readily accessible ECG signals, which are highly coherent with ECHO videos, bypassing complex conditional prompts. To the best of our knowledge, this is the first work to use time-series prompts like ECG signals for ECHO video generation. ECHOPULSE not only enables controllable synthetic ECHO data generation but also provides updated cardiac function information for disease monitoring and prediction beyond ECG alone. Evaluations on three public and private datasets demonstrate state-of-the-art performance in ECHO video generation across both qualitative and quantitative measures. Additionally, ECHOPULSE can be easily generalized to other modality generation tasks, such as cardiac MRI, fMRI, and 3D CT generation. Demo can seen from \url{https://github.com/levyisthebest/ECHOPulse_Prelease}.

This book, Design Patterns in Machine Learning and Deep Learning: Advancing Big Data Analytics Management, presents a comprehensive study of essential design patterns tailored for large-scale machine learning and deep learning applications. The book explores the application of classical software engineering patterns, Creational, Structural, Behavioral, and Concurrency Patterns, to optimize the development, maintenance, and scalability of big data analytics systems. Through practical examples and detailed Python implementations, it bridges the gap between traditional object-oriented design patterns and the unique demands of modern data analytics environments. Key design patterns such as Singleton, Factory, Observer, and Strategy are analyzed for their impact on model management, deployment strategies, and team collaboration, providing invaluable insights into the engineering of efficient, reusable, and flexible systems. This volume is an essential resource for developers, researchers, and engineers aiming to enhance their technical expertise in both machine learning and software design., Comment: 138pages

Optical phonon engineering through nonlinear effects has been utilized in ultrafast control of material properties. However, nonlinear optical phonons typically exhibit rapid decay due to strong mode-mode couplings, limiting their effectiveness in temperature or frequency sensitive applications. In this study, we report the observation of long-lived nonlinear optical phonons through the spontaneous formation of phonon frequency combs in the van der Waals material CrXTe$_3$ (X=Ge, Si) using high-resolution Raman scattering. Unlike conventional optical phonons, the highest $A_g$ mode in CrGeTe$_3$ splits into equidistant, sharp peaks forming a frequency comb that persists for hundreds of oscillations and survives up to 100K before decaying. These modes correspond to localized oscillations of Ge$_2$Te$_6$ clusters, isolated from Cr hexagons, behaving as independent quantum oscillators. Introducing a cubic nonlinear term to the harmonic oscillator model, we simulate the phonon time evolution and successfully replicate the observed comb structure. Similar frequency comb behavior is observed in CrSiTe$_3$, demonstrating the generalizability of this phenomenon. Our findings reveal that Raman scattering effectively probes high-frequency nonlinear phonon modes, providing new insight into generating long-lived, tunable phonon frequency combs with applications in ultrafast material control and phonon-based technologies., Comment: 22 pages, 10 figures

This paper explores the integration of ring attractors, a mathematical model inspired by neural circuit dynamics, into the reinforcement learning (RL) action selection process. Ring attractors, as specialized brain-inspired structures that encode spatial information and uncertainty, offer a biologically plausible mechanism to improve learning speed and predictive performance. They do so by explicitly encoding the action space, facilitating the organization of neural activity, and enabling the distribution of spatial representations across the neural network in the context of deep RL. The application of ring attractors in the RL action selection process involves mapping actions to specific locations on the ring and decoding the selected action based on neural activity. We investigate the application of ring attractors by both building them as exogenous models and integrating them as part of a Deep Learning policy algorithm. Our results show a significant improvement in state-of-the-art models for the Atari 100k benchmark. Notably, our integrated approach improves the performance of state-of-the-art models by half, representing a 53\% increase over selected baselines.

Although flares from late-type main-sequence stars have been frequently detected in multi-wavelength, the associated dynamical process has been rarely reported so far. Here, we report follow-up observations of an X-ray transient triggered by WXT onboard the Einstein Probe at UT08:45:08 in 2024, May 7. The photometry in multi-bands and time-resolved spectroscopy started at 3 and 7.5 hours after the trigger, respectively, which enables us to identify the transient as a flare of the M-dwarf 2MASS J12184187-0609123. The bolometric energy released in the flare is estimated to be $\sim10^{36}\ \mathrm{erg}$ from its X-ray light curve. The H$\alpha$ emission-line profile obtained at about 7 hours after the trigger shows an evident blue asymmetry with a maximum velocity of $200-250\ \mathrm{km\ s^{-1}}$. The blue wing can be likely explained by the chromospheric temperature (cool) upflow associated with chromospheric evaporation, in which the mass of the evaporating plasma is estimated to be $1.2\times10^{18}$g. In addition, a prominence eruption with an estimated mass of $7\times10^{15}\mathrm{g}

Dissipation is usually considered a negative factor for observing quantum effects and for harnessing them for quantum technologies. Here we propose a scheme for speeding up the generation of quantum entanglement between two coupled qubits by introducing a strong dissipation channel to one of these qubits. The maximal entanglement is conditionally established by evenly distributing a single excitation between these two qubits. When the excitation is initially held by the dissipative qubit, the dissipation accelerates the excitation re-distribution process for the quantum state trajectory without quantum jumps. Our results show that the time needed to conditionally attain the maximal entanglement is monotonously decreased as the dissipative rate is increased. We further show that this scheme can be generalized to accelerate the production of the W state for the three-qubit system, where one NH qubit is symmetrically coupled to two Hermitian qubits., Comment: 2 figures

Object navigation in unknown environments is crucial for deploying embodied agents in real-world applications. While we have witnessed huge progress due to large-scale scene datasets, faster simulators, and stronger models, previous studies mainly focus on limited scene types and target objects. In this paper, we study a new task of navigating to diverse target objects in a large number of scene types. To benchmark the problem, we present a large-scale scene dataset, DivScene, which contains 4,614 scenes across 81 different types. With the dataset, we build an end-to-end embodied agent, NatVLM, by fine-tuning a Large Vision Language Model (LVLM) through imitation learning. The LVLM is trained to take previous observations from the environment and generate the next actions. We also introduce CoT explanation traces of the action prediction for better performance when tuning LVLMs. Our extensive experiments find that we can build a performant LVLM-based agent through imitation learning on the shortest paths constructed by a BFS planner without any human supervision. Our agent achieves a success rate that surpasses GPT-4o by over 20%. Meanwhile, we carry out various analyses showing the generalization ability of our agent. Our code and data are available at https://github.com/zhaowei-wang-nlp/DivScene., Comment: Work in Progress

Accurate 3D object detection in real-world environments requires a huge amount of annotated data with high quality. Acquiring such data is tedious and expensive, and often needs repeated effort when a new sensor is adopted or when the detector is deployed in a new environment. We investigate a new scenario to construct 3D object detectors: learning from the predictions of a nearby unit that is equipped with an accurate detector. For example, when a self-driving car enters a new area, it may learn from other traffic participants whose detectors have been optimized for that area. This setting is label-efficient, sensor-agnostic, and communication-efficient: nearby units only need to share the predictions with the ego agent (e.g., car). Naively using the received predictions as ground-truths to train the detector for the ego car, however, leads to inferior performance. We systematically study the problem and identify viewpoint mismatches and mislocalization (due to synchronization and GPS errors) as the main causes, which unavoidably result in false positives, false negatives, and inaccurate pseudo labels. We propose a distance-based curriculum, first learning from closer units with similar viewpoints and subsequently improving the quality of other units' predictions via self-training. We further demonstrate that an effective pseudo label refinement module can be trained with a handful of annotated data, largely reducing the data quantity necessary to train an object detector. We validate our approach on the recently released real-world collaborative driving dataset, using reference cars' predictions as pseudo labels for the ego car. Extensive experiments including several scenarios (e.g., different sensors, detectors, and domains) demonstrate the effectiveness of our approach toward label-efficient learning of 3D perception from other units' predictions.

Lifelong user behavior sequences, comprising up to tens of thousands of history behaviors, are crucial for capturing user interests and predicting user responses in modern recommendation systems. A two-stage paradigm is typically adopted to handle these long sequences: a few relevant behaviors are first searched from the original long sequences via an attention mechanism in the first stage and then aggregated with the target item to construct a discriminative representation for prediction in the second stage. In this work, we identify and characterize, for the first time, a neglected deficiency in existing long-sequence recommendation models: a single set of embeddings struggles with learning both attention and representation, leading to interference between these two processes. Initial attempts to address this issue using linear projections -- a technique borrowed from language processing -- proved ineffective, shedding light on the unique challenges of recommendation models. To overcome this, we propose the Decoupled Attention and Representation Embeddings (DARE) model, where two distinct embedding tables are initialized and learned separately to fully decouple attention and representation. Extensive experiments and analysis demonstrate that DARE provides more accurate search of correlated behaviors and outperforms baselines with AUC gains up to 0.9% on public datasets and notable online system improvements. Furthermore, decoupling embedding spaces allows us to reduce the attention embedding dimension and accelerate the search procedure by 50% without significant performance impact, enabling more efficient, high-performance online serving., Comment: First three authors contributed equally

Recently, 3D Gaussian Splatting (3DGS) has exceled in novel view synthesis with its real-time rendering capabilities and superior quality. However, it faces challenges for high-resolution novel view synthesis (HRNVS) due to the coarse nature of primitives derived from low-resolution input views. To address this issue, we propose Super-Resolution 3DGS (SuperGS), which is an expansion of 3DGS designed with a two-stage coarse-to-fine training framework, utilizing pretrained low-resolution scene representation as an initialization for super-resolution optimization. Moreover, we introduce Multi-resolution Feature Gaussian Splatting (MFGS) to incorporates a latent feature field for flexible feature sampling and Gradient-guided Selective Splitting (GSS) for effective Gaussian upsampling. By integrating these strategies within the coarse-to-fine framework ensure both high fidelity and memory efficiency. Extensive experiments demonstrate that SuperGS surpasses state-of-the-art HRNVS methods on challenging real-world datasets using only low-resolution inputs.

We introduce ColaCare, a framework that enhances Electronic Health Record (EHR) modeling through multi-agent collaboration driven by Large Language Models (LLMs). Our approach seamlessly integrates domain-specific expert models with LLMs to bridge the gap between structured EHR data and text-based reasoning. Inspired by clinical consultations, ColaCare employs two types of agents: DoctorAgent and MetaAgent, which collaboratively analyze patient data. Expert models process and generate predictions from numerical EHR data, while LLM agents produce reasoning references and decision-making reports within the collaborative consultation framework. We additionally incorporate the Merck Manual of Diagnosis and Therapy (MSD) medical guideline within a retrieval-augmented generation (RAG) module for authoritative evidence support. Extensive experiments conducted on four distinct EHR datasets demonstrate ColaCare's superior performance in mortality prediction tasks, underscoring its potential to revolutionize clinical decision support systems and advance personalized precision medicine. The code, complete prompt templates, more case studies, etc. are publicly available at the anonymous link: https://colacare.netlify.app.

Many people struggle with learning a new language, with traditional tools falling short in providing contextualized learning tailored to each learner's needs. The recent development of large language models (LLMs) and embodied conversational agents (ECAs) in social virtual reality (VR) provide new opportunities to practice language learning in a contextualized and naturalistic way that takes into account the learner's language level and needs. To explore this opportunity, we developed ELLMA-T, an ECA that leverages an LLM (GPT-4) and situated learning framework for supporting learning English language in social VR (VRChat). Drawing on qualitative interviews (N=12), we reveal the potential of ELLMA-T to generate realistic, believable and context-specific role plays for agent-learner interaction in VR, and LLM's capability to provide initial language assessment and continuous feedback to learners. We provide five design implications for the future development of LLM-based language agents in social VR., Comment: 20 pages, 6 figures

Deep neural network based systems deployed in sensitive environments are vulnerable to adversarial attacks. Unrestricted adversarial attacks typically manipulate the semantic content of an image (e.g., color or texture) to create adversarial examples that are both effective and photorealistic. Recent works have utilized the diffusion inversion process to map images into a latent space, where high-level semantics are manipulated by introducing perturbations. However, they often results in substantial semantic distortions in the denoised output and suffers from low efficiency. In this study, we propose a novel framework called Semantic-Consistent Unrestricted Adversarial Attacks (SCA), which employs an inversion method to extract edit-friendly noise maps and utilizes Multimodal Large Language Model (MLLM) to provide semantic guidance throughout the process. Under the condition of rich semantic information provided by MLLM, we perform the DDPM denoising process of each step using a series of edit-friendly noise maps, and leverage DPM Solver++ to accelerate this process, enabling efficient sampling with semantic consistency. Compared to existing methods, our framework enables the efficient generation of adversarial examples that exhibit minimal discernible semantic changes. Consequently, we for the first time introduce Semantic-Consistent Adversarial Examples (SCAE). Extensive experiments and visualizations have demonstrated the high efficiency of SCA, particularly in being on average 12 times faster than the state-of-the-art attacks. Our research can further draw attention to the security of multimedia information.

Multi-objective reinforcement learning (MORL) excels at handling rapidly changing preferences in tasks that involve multiple criteria, even for unseen preferences. However, previous dominating MORL methods typically generate a fixed policy set or preference-conditioned policy through multiple training iterations exclusively for sampled preference vectors, and cannot ensure the efficient discovery of the Pareto front. Furthermore, integrating preferences into the input of policy or value functions presents scalability challenges, in particular as the dimension of the state and preference space grow, which can complicate the learning process and hinder the algorithm's performance on more complex tasks. To address these issues, we propose a two-stage Pareto front discovery algorithm called Constrained MORL (C-MORL), which serves as a seamless bridge between constrained policy optimization and MORL. Concretely, a set of policies is trained in parallel in the initialization stage, with each optimized towards its individual preference over the multiple objectives. Then, to fill the remaining vacancies in the Pareto front, the constrained optimization steps are employed to maximize one objective while constraining the other objectives to exceed a predefined threshold. Empirically, compared to recent advancements in MORL methods, our algorithm achieves more consistent and superior performances in terms of hypervolume, expected utility, and sparsity on both discrete and continuous control tasks, especially with numerous objectives (up to nine objectives in our experiments)., Comment: 27 pages, 8 figues. In Submission to a conference

Hierarchical Imitation Learning (HIL) is a promising approach for tackling long-horizon decision-making tasks. While it is a challenging task due to the lack of detailed supervisory labels for sub-goal learning, and reliance on hundreds to thousands of expert demonstrations. In this work, we introduce SEAL, a novel framework that leverages Large Language Models (LLMs)'s powerful semantic and world knowledge for both specifying sub-goal space and pre-labeling states to semantically meaningful sub-goal representations without prior knowledge of task hierarchies. SEAL employs a dual-encoder structure, combining supervised LLM-guided sub-goal learning with unsupervised Vector Quantization (VQ) for more robust sub-goal representations. Additionally, SEAL incorporates a transition-augmented low-level planner for improved adaptation to sub-goal transitions. Our experiments demonstrate that SEAL outperforms state-of-the-art HIL methods and LLM-based planning approaches, particularly in settings with small expert datasets and complex long-horizon tasks., Comment: 18 pages, 5 figures, in submission

Sexual harassment has been recognized as a significant social issue. In recent years, the emergence of harassment in social virtual reality (VR) has become an important and urgent research topic. We employed a mixed-methods approach by conducting online surveys with VR users (N = 166) and semi-structured interviews with social VR users (N = 18) to investigate how users perceive sexual harassment in social VR, focusing on the influence of avatar appearance. Moreover, we derived users' response strategies to sexual harassment and gained insights on platform regulation. This study contributes to the research on sexual harassment in social VR by examining the moderating effect of avatar appearance on user perception of sexual harassment and uncovering the underlying reasons behind response strategies. Moreover, it presents novel prospects and challenges in platform design and regulation domains.

Code completion, a key downstream task in code generation, is one of the most frequent and impactful methods for enhancing developer productivity in software development. As intelligent completion tools evolve, we need a robust evaluation benchmark that enables meaningful comparisons between products and guides future advancements. However, existing benchmarks focus more on coarse-grained tasks without industrial analysis resembling general code generation rather than the real-world scenarios developers encounter. Moreover, these benchmarks often rely on costly and time-consuming human annotation, and the standalone test cases fail to leverage minimal tests for maximum repository-level understanding and code coverage. To address these limitations, we first analyze business data from an industrial code completion tool and redefine the evaluation criteria to better align with the developer's intent and desired completion behavior throughout the coding process. Based on these insights, we introduce Codev-Agent, an agent-based system that automates repository crawling, constructs execution environments, extracts dynamic calling chains from existing unit tests, and generates new test samples to avoid data leakage, ensuring fair and effective comparisons. Using Codev-Agent, we present the Code-Development Benchmark (Codev-Bench), a fine-grained, real-world, repository-level, and developer-centric evaluation framework. Codev-Bench assesses whether a code completion tool can capture a developer's immediate intent and suggest appropriate code across diverse contexts, providing a more realistic benchmark for code completion in modern software development.

Zero-shot voice conversion (VC) aims to transform the source speaker timbre into an arbitrary unseen one without altering the original speech content.While recent advancements in zero-shot VC methods have shown remarkable progress, there still remains considerable potential for improvement in terms of improving speaker similarity and speech naturalness.In this paper, we propose Takin-VC, a novel zero-shot VC framework based on jointly hybrid content and memory-augmented context-aware timbre modeling to tackle this challenge. Specifically, an effective hybrid content encoder, guided by neural codec training, that leverages quantized features from pre-trained WavLM and HybridFormer is first presented to extract the linguistic content of the source speech. Subsequently, we introduce an advanced cross-attention-based context-aware timbre modeling approach that learns the fine-grained, semantically associated target timbre features. To further enhance both speaker similarity and real-time performance, we utilize a conditional flow matching model to reconstruct the Mel-spectrogram of the source speech. Additionally, we advocate an efficient memory-augmented module designed to generate high-quality conditional target inputs for the flow matching process, thereby improving the overall performance of the proposed system. Experimental results demonstrate that the proposed Takin-VC method surpasses state-of-the-art zero-shot VC systems, delivering superior performance in terms of both speech naturalness and speaker similarity., Comment: Work in Progress; Under Review

This book serves as an introduction to deep learning and machine learning, focusing on their applications in big data analytics. It covers essential concepts, tools like ChatGPT and Claude, hardware recommendations, and practical guidance on setting up development environments using libraries like PyTorch and TensorFlow. Designed for beginners and advanced users alike, it provides step-by-step instructions, hands-on projects, and insights into AI's future, including AutoML and edge computing., Comment: This book contains 156 pages and 9 figures

It is generally accepted that the evolution of deep-water surface gravity wave spectrum is governed by quartet resonant and quasi-resonant interactions. However, it has also been reported in both experimental and computational studies that non-resonant triad interactions can play a role, e.g., generation of bound waves. In this study, we investigate the effects of triad and quartet interactions on the spectral evolution, by numerically tracking the contributions from quadratic and cubic terms in the dynamical equation. In a finite time interval, we find that the contribution from triad interactions follows the trend of that from quartet resonances (with comparable magnitude) for most wavenumbers, except that it peaks at low wavenumbers with very low initial energy. This result reveals two effects of triad interactions: (1) the non-resonant triad interactions can be connected to form quartet resonant interactions (hence exhibiting the comparable trend), which is a reflection of the normal form transformation applied in wave turbulence theory of surface gravity waves. (2) the triad interactions can fill energy into the low energy portion of the spectrum (low wavenumber part in this case) on a very fast time scale, with energy distributed in both bound and free modes at the same wavenumber. We further analyze the latter mechanism using a simple model with two initially active modes in the wavenumber domain. Analytical formulae are provided to describe the distribution of energy in free and bound modes with numerical validations.