Search

Sound Event Detection (SED) detects regions of sound events, while Speaker Diarization (SD) segments speech conversations attributed to individual speakers. In SED, all speaker segments are classified as a single speech event, while in SD, non-speech sounds are treated merely as background noise. Thus, both tasks provide only partial analysis in complex audio scenarios involving both speech conversation and non-speech sounds. In this paper, we introduce a novel task called Unified Audio Event Detection (UAED) for comprehensive audio analysis. UAED explores the synergy between SED and SD tasks, simultaneously detecting non-speech sound events and fine-grained speech events based on speaker identities. To tackle this task, we propose a Transformer-based UAED (T-UAED) framework and construct the UAED Data derived from the Librispeech dataset and DESED soundbank. Experiments demonstrate that the proposed framework effectively exploits task interactions and substantially outperforms the baseline that simply combines the outputs of SED and SD models. T-UAED also shows its versatility by performing comparably to specialized models for individual SED and SD tasks on DESED and CALLHOME datasets., Comment: submitted to ICASSP 2025

Novel-view synthesis based on visible light has been extensively studied. In comparison to visible light imaging, thermal infrared imaging offers the advantage of all-weather imaging and strong penetration, providing increased possibilities for reconstruction in nighttime and adverse weather scenarios. However, thermal infrared imaging is influenced by physical characteristics such as atmospheric transmission effects and thermal conduction, hindering the precise reconstruction of intricate details in thermal infrared scenes, manifesting as issues of floaters and indistinct edge features in synthesized images. To address these limitations, this paper introduces a physics-induced 3D Gaussian splatting method named Thermal3D-GS. Thermal3D-GS begins by modeling atmospheric transmission effects and thermal conduction in three-dimensional media using neural networks. Additionally, a temperature consistency constraint is incorporated into the optimization objective to enhance the reconstruction accuracy of thermal infrared images. Furthermore, to validate the effectiveness of our method, the first large-scale benchmark dataset for this field named Thermal Infrared Novel-view Synthesis Dataset (TI-NSD) is created. This dataset comprises 20 authentic thermal infrared video scenes, covering indoor, outdoor, and UAV(Unmanned Aerial Vehicle) scenarios, totaling 6,664 frames of thermal infrared image data. Based on this dataset, this paper experimentally verifies the effectiveness of Thermal3D-GS. The results indicate that our method outperforms the baseline method with a 3.03 dB improvement in PSNR and significantly addresses the issues of floaters and indistinct edge features present in the baseline method. Our dataset and codebase will be released in \href{https://github.com/mzzcdf/Thermal3DGS}{\textcolor{red}{Thermal3DGS}}., Comment: 17 pages, 4 figures, 3 tables

Advances in lithium-metal anodes have inspired interest in discovery of Li-free cathodes, most of which are natively found in their charged state. This is in contrast to today's commercial lithium-ion battery cathodes, which are more stable in their discharged state. In this study, we combine calculated cathode voltage information from both categories of cathode materials, covering 5577 and 2423 total unique structure pairs, respectively. The resulting voltage distributions with respect to the redox pairs and anion types for both classes of compounds emphasize design principles for high-voltage cathodes, which favor later Period 4 transition metals in their higher oxidation states and more electronegative anions like fluorine or polyaion groups. Generally, cathodes that are found in their charged, delithiated state are shown to exhibit voltages lower than those that are most stable in their lithiated state, in agreement with thermodynamic expectations. Deviations from this trend are found to originate from different anion distributions between redox pairs. In addition, a machine learning model for voltage prediction based on chemical formulae is constructed, and shows state-of-the-art performance when compared to two established composition-based ML models for materials properties predictions, Roost and CrabNet.

Most star formation in the local Universe occurs in spiral galaxies, but their origin remains an unanswered question. Various theories have been proposed to explain the development of spiral arms, each predicting different spatial distributions of the interstellar medium. This study maps the star formation rate (SFR) and gas-phase metallicity of nine spiral galaxies with the TYPHOON survey to test two dominating theories: density wave theory and dynamic spiral theory. We discuss the environmental effects on our galaxies, considering reported environments and merging events. Taking advantage of the large field of view covering the entire optical disk, we quantify the fluctuation of SFR and metallicity relative to the azimuthal distance from the spiral arms. We find higher SFR and metallicity in the trailing edge of NGC~1365 (by 0.117~dex and 0.068~dex, respectively) and NGC~1566 (by 0.119~dex and 0.037~dex, respectively), which is in line with density wave theory. NGC~2442 shows a different result with higher metallicity (0.093~dex) in the leading edge, possibly attributed to an ongoing merging. The other six spiral galaxies show no statistically significant offset in SFR or metallicity, consistent with dynamic spiral theory. We also compare the behaviour of metallicity inside and outside the co-rotation radius (CR) of NGC~1365 and NGC~1566. We find comparable metallicity fluctuations near and beyond the CR of NGC~1365, indicating gravitational perturbation. NGC~1566 shows the greatest fluctuation near the CR, in line with the analytic spiral arms. Our work highlights that a combination of mechanisms explains the origin of spiral features in the local Universe., Comment: 22 pages, 10 figures, 4 tables, accepted for publication in MNRAS

To achieve truly seamless global intelligent connectivity, non-terrestrial networks (NTN) mainly composed of low earth orbit (LEO) satellites and drones are recognized as important components of the future 6G network architecture. Meanwhile, the rapid advancement of the Internet of Things (IoT) has led to the proliferation of numerous applications with stringent requirements for timely information delivery. The Age of Information (AoI), a critical performance metric for assessing the freshness of data in information update systems, has gained significant importance in this context. However, existing modeling and analysis work on AoI mainly focuses on terrestrial networks, and the distribution characteristics of ground nodes and the high dynamics of satellites have not been fully considered, which poses challenges for more accurate evaluation. Against this background, we model the ground nodes as a hybrid distribution of Poisson point process (PPP) and Poisson cluster process (PCP) to capture the impact of ground node distribution on the AoI of status update packet transmission supported by UAVs and satellites in NTN, and the visibility and cross-traffic characteristics of satellites are additionally considered. We derived the average AoI for the system in these two different situations and examined the impact of various network parameters on AoI performance., Comment: 6 pages, 5 figures

Language models have been effectively applied to modeling natural signals, such as images, video, speech, and audio. A crucial component of these models is the codec tokenizer, which compresses high-dimensional natural signals into lower-dimensional discrete tokens. In this paper, we introduce WavTokenizer, which offers several advantages over previous SOTA acoustic codec models in the audio domain: 1)extreme compression. By compressing the layers of quantizers and the temporal dimension of the discrete codec, one-second audio of 24kHz sampling rate requires only a single quantizer with 40 or 75 tokens. 2)improved subjective quality. Despite the reduced number of tokens, WavTokenizer achieves state-of-the-art reconstruction quality with outstanding UTMOS scores and inherently contains richer semantic information. Specifically, we achieve these results by designing a broader VQ space, extended contextual windows, and improved attention networks, as well as introducing a powerful multi-scale discriminator and an inverse Fourier transform structure. We conducted extensive reconstruction experiments in the domains of speech, audio, and music. WavTokenizer exhibited strong performance across various objective and subjective metrics compared to state-of-the-art models. We also tested semantic information, VQ utilization, and adaptability to generative models. Comprehensive ablation studies confirm the necessity of each module in WavTokenizer. The related code, demos, and pre-trained models are available at https://github.com/jishengpeng/WavTokenizer., Comment: Working in progress. arXiv admin note: text overlap with arXiv:2402.12208

Speaker diarization, the process of segmenting an audio stream or transcribed speech content into homogenous partitions based on speaker identity, plays a crucial role in the interpretation and analysis of human speech. Most existing speaker diarization systems rely exclusively on unimodal acoustic information, making the task particularly challenging due to the innate ambiguities of audio signals. Recent studies have made tremendous efforts towards audio-visual or audio-semantic modeling to enhance performance. However, even the incorporation of up to two modalities often falls short in addressing the complexities of spontaneous and unstructured conversations. To exploit more meaningful dialogue patterns, we propose a novel multimodal approach that jointly utilizes audio, visual, and semantic cues to enhance speaker diarization. Our method elegantly formulates the multimodal modeling as a constrained optimization problem. First, we build insights into the visual connections among active speakers and the semantic interactions within spoken content, thereby establishing abundant pairwise constraints. Then we introduce a joint pairwise constraint propagation algorithm to cluster speakers based on these visual and semantic constraints. This integration effectively leverages the complementary strengths of different modalities, refining the affinity estimation between individual speaker embeddings. Extensive experiments conducted on multiple multimodal datasets demonstrate that our approach consistently outperforms state-of-the-art speaker diarization methods.

Automatic Speech Recognition (ASR) transcripts exhibit recognition errors and various spoken language phenomena such as disfluencies, ungrammatical sentences, and incomplete sentences, hence suffering from poor readability. To improve readability, we propose a Contextualized Spoken-to-Written conversion (CoS2W) task to address ASR and grammar errors and also transfer the informal text into the formal style with content preserved, utilizing contexts and auxiliary information. This task naturally matches the in-context learning capabilities of Large Language Models (LLMs). To facilitate comprehensive comparisons of various LLMs, we construct a document-level Spoken-to-Written conversion of ASR Transcripts Benchmark (SWAB) dataset. Using SWAB, we study the impact of different granularity levels on the CoS2W performance, and propose methods to exploit contexts and auxiliary information to enhance the outputs. Experimental results reveal that LLMs have the potential to excel in the CoS2W task, particularly in grammaticality and formality, our methods achieve effective understanding of contexts and auxiliary information by LLMs. We further investigate the effectiveness of using LLMs as evaluators and find that LLM evaluators show strong correlations with human evaluations on rankings of faithfulness and formality, which validates the reliability of LLM evaluators for the CoS2W task., Comment: 7 pages, 3 figures

We investigate the physical properties of steady flows in a holographic first-order phase transition model, extending from the thermodynamics at equilibrium to the real-time dynamics far from equilibrium. Through spinodal decomposition or condensation nuclei, the phase-separated state with non-zero momentum can be achieved. In this scenario, we observe a gap between coexisting phases, arising not only from the variations in energy density, but also from the distinctions in momentum density or longitudinal pressure. These disparities are characterized by flow velocity and latent heat. Furthermore, by introducing an inhomogeneous scalar external source to simulate a fixed obstacle, we reveal the dynamical response of momentum loss in the moving system. Notably, starting from an initial phase-separated state with uniform flow velocity, and subsequently interacting it with an obstacle, we find that the moving high-energy phase exhibits four characteristic dynamical behaviors -- rebounding, pinning, passing, and splitting. These behaviors depend on the velocity of the phase and the strength of the obstacle.

The video topic segmentation (VTS) task segments videos into intelligible, non-overlapping topics, facilitating efficient comprehension of video content and quick access to specific content. VTS is also critical to various downstream video understanding tasks. Traditional VTS methods using shallow features or unsupervised approaches struggle to accurately discern the nuances of topical transitions. Recently, supervised approaches have achieved superior performance on video action or scene segmentation over unsupervised approaches. In this work, we improve supervised VTS by thoroughly exploring multimodal fusion and multimodal coherence modeling. Specifically, (1) we enhance multimodal fusion by exploring different architectures using cross-attention and mixture of experts. (2) To generally strengthen multimodality alignment and fusion, we pre-train and fine-tune the model with multimodal contrastive learning. (3) We propose a new pre-training task tailored for the VTS task, and a novel fine-tuning task for enhancing multimodal coherence modeling for VTS. We evaluate the proposed approaches on educational videos, in the form of lectures, due to the vital role of topic segmentation of educational videos in boosting learning experiences. Additionally, we introduce a large-scale Chinese lecture video dataset to augment the existing English corpus, promoting further research in VTS. Experiments on both English and Chinese lecture datasets demonstrate that our model achieves superior VTS performance compared to competitive unsupervised and supervised baselines.

Encoding frequency stability constraints in the operation problem is challenging due to its complex dynamics. Recently, data-driven approaches have been proposed to learn the stability criteria offline with the trained model embedded as a constraint of online optimization. However, random sampling of stationary operation points is less efficient in generating balanced stable and unstable samples. Meanwhile, the performance of such a model is strongly dependent on the quality of the training dataset. Observing this research gap, we propose a gradient-based data generation method via forward-mode automatic differentiation. In this method, the original dynamic system is augmented with new states that represent the dynamic of sensitivities of the original states, which can be solved by invoking any ODE solver for a single time. To compensate for the contradiction between the gradient of various frequency stability criteria, gradient surgery is proposed by projecting the gradient on the normal plane of the other. In the end, we demonstrate the superior performance of the proposed sampling algorithm, compared with the unrolling differentiation and finite difference. All codes are available at https://github.com/xuwkk/frequency_sample_ad.

High-quality 3D photoacoustic imaging (PAI) reconstruction under sparse view or limited view has long been challenging. Traditional 3D iterative-based reconstruction methods suffer from both slow speed and high memory consumption. Recently, in computer graphics, the differentiable rendering has made significant progress, particularly with the rise of 3D Gaussian Splatting. Inspired by these, we introduce differentiable radiation into PAI, developing a novel reconstruction algorithm: the Sliding Ball Adaptive Growth algorithm (SlingBAG) for 3D PAI, which shows ability in high-quality 3D PAI reconstruction both under extremely sparse view and limited view. We established the point cloud dataset in PAI, and used unique differentiable rapid radiator based on the spherical decomposition strategy and the randomly initialized point cloud adaptively optimized according to sparse sensor data. Each point undergoes updates in 3D coordinates, initial pressure, and resolution (denoted by the radius of ball). Points undergo adaptive growth during iterative process, including point destroying, splitting and duplicating along the gradient of their positions, manifesting the sliding ball effect. Finally, our point cloud to voxel grid shader renders the final reconstruction results. Simulation and in vivo experiments demonstrate that our SlingBAG reconstruction result's SNR can be more than 40 dB under extremely sparse view, while the SNR of traditional back-projection algorithm's result is less than 20 dB. Moreover, the result of SlingBAG's structural similarity to the ground truth is significantly higher, with an SSIM value of 95.6%. Notably, our differentiable rapid radiator can conduct forward PA simulation in homogeneous, non-viscous media substantially faster than current methods that numerically simulate the wave propagation, such as k-Wave. The dataset and all code will be open source.

Many small and medium-sized physics experiments are being conducted worldwide. These experiments have similar requirements for readout electronics, especially the back-end electronics. Some experiments need a trigger logic unit(TLU) to provide timing and synchronous control signals. This paper introduces a back-end electronics design for small and medium-sized physics experiments; it adopts a daughter-motherboard structure integrated TLU function to provide greater flexibility. Different interfaces and protocols can be flexibly selected according to data bandwidth requirements. It supports 32 optical fiber interfaces based on a field-programmable gate array (FPGA) of normal IOs with 400Mbps of data bandwidth for each channel. At the same time, it supports 16 high-speed communication interfaces based on GTX port with several Gbps data bandwidth of each channel. For the TLU function, this design has 8 HDMI interfaces and one RJ45 interface to provide synchronous triggers and other control signals, and it has six analog LEMOs and four digital LEMOs to accept asynchronous signals from an external source. These design specifications can meet the needs of most small and medium-sized experiments. This set of back-end electronics has been successfully used in experiments such as PandaX-III, VLAST, and moungraphy. Moreover, it has successfully conducted beam tests with a readout of the data of VLAST detectors at CERN.

The metal content of galaxies is a direct probe of the baryon cycle. A hallmark example is the relationship between a galaxy's stellar mass, star formation rate (SFR), and gas-phase metallicity: the Fundamental Metallicity Relation (FMR). While low-redshift ($z\lesssim4$) observational studies suggest that the FMR is redshift-invariant, recent JWST data indicate deviations from this model. In this study, we utilize the FMR to predict the evolution of the normalisation of the mass-metallicity relation (MZR) using the cosmological simulations Illustris, IllustrisTNG, EAGLE, and SIMBA. Our findings demonstrate that a $z = 0$ calibrated FMR struggles to predict the evolution in the MZR of each simulation. To quantify the divergence of the predictions, we introduce the concepts of a ''static'' FMR, where the role of the SFR in setting the normalization of the MZR does not change with redshift, and a ''dynamic'' FMR, where the role of SFR evolves over time. We find static FMRs in Illustris and SIMBA and dynamic FMRs in IllustrisTNG and EAGLE. We suggest that the differences between these models likely points to the subtle differences in the implementation of the baryon cycle. Moreover, we echo recent JWST results at $z > 4$ by finding significant offsets from the FMR in IllustrisTNG and EAGLE, suggesting that the observed FMR may be dynamic as well. Overall, our findings imply that the current FMR framework neglects important variations in the baryon cycle through cosmic time., Comment: Submitted to MNRAS. 17 pages, 8 Figures. See also paper one of series (arXiv:2403.08856). Comments welcome

Recent years have witnessed a trend that large language model (LLM) based text-to-speech (TTS) emerges into the mainstream due to their high naturalness and zero-shot capacity. In this paradigm, speech signals are discretized into token sequences, which are modeled by an LLM with text as prompts and reconstructed by a token-based vocoder to waveforms. Obviously, speech tokens play a critical role in LLM-based TTS models. Current speech tokens are learned in an unsupervised manner, which lacks explicit semantic information and alignment to the text. In this paper, we propose to represent speech with supervised semantic tokens, which are derived from a multilingual speech recognition model by inserting vector quantization into the encoder. Based on the tokens, we further propose a scalable zero-shot TTS synthesizer, CosyVoice, which consists of an LLM for text-to-token generation and a conditional flow matching model for token-to-speech synthesis. Experimental results show that supervised semantic tokens significantly outperform existing unsupervised tokens in terms of content consistency and speaker similarity for zero-shot voice cloning. Moreover, we find that utilizing large-scale data further improves the synthesis performance, indicating the scalable capacity of CosyVoice. To the best of our knowledge, this is the first attempt to involve supervised speech tokens into TTS models., Comment: work in progress. arXiv admin note: substantial text overlap with arXiv:2407.04051

This report introduces FunAudioLLM, a model family designed to enhance natural voice interactions between humans and large language models (LLMs). At its core are two innovative models: SenseVoice, which handles multilingual speech recognition, emotion recognition, and audio event detection; and CosyVoice, which facilitates natural speech generation with control over multiple languages, timbre, speaking style, and speaker identity. SenseVoice-Small delivers exceptionally low-latency ASR for 5 languages, and SenseVoice-Large supports high-precision ASR for over 50 languages, while CosyVoice excels in multi-lingual voice generation, zero-shot in-context learning, cross-lingual voice cloning, and instruction-following capabilities. The models related to SenseVoice and CosyVoice have been open-sourced on Modelscope and Huggingface, along with the corresponding training, inference, and fine-tuning codes released on GitHub. By integrating these models with LLMs, FunAudioLLM enables applications such as speech-to-speech translation, emotional voice chat, interactive podcasts, and expressive audiobook narration, thereby pushing the boundaries of voice interaction technology. Demos are available at https://fun-audio-llm.github.io, and the code can be accessed at https://github.com/FunAudioLLM., Comment: Work in progress. Authors are listed in alphabetical order by family name

Large Language Models (LLMs) have showcased exceptional capabilities in various domains, attracting significant interest from both academia and industry. Despite their impressive performance, the substantial size and computational demands of LLMs pose considerable challenges for practical deployment, particularly in environments with limited resources. The endeavor to compress language models while maintaining their accuracy has become a focal point of research. Among the various methods, knowledge distillation has emerged as an effective technique to enhance inference speed without greatly compromising performance. This paper presents a thorough survey from three aspects: method, evaluation, and application, exploring knowledge distillation techniques tailored specifically for LLMs. Specifically, we divide the methods into white-box KD and black-box KD to better illustrate their differences. Furthermore, we also explored the evaluation tasks and distillation effects between different distillation methods, and proposed directions for future research. Through in-depth understanding of the latest advancements and practical applications, this survey provides valuable resources for researchers, paving the way for sustained progress in this field., Comment: 28 pages

LLMs have achieved significant performance progress in various NLP applications. However, LLMs still struggle to meet the strict requirements for accuracy and reliability in the medical field and face many challenges in clinical applications. Existing clinical diagnostic evaluation benchmarks for evaluating medical agents powered by LLMs have severe limitations. Firstly, most existing medical evaluation benchmarks face the risk of data leakage or contamination. Secondly, existing benchmarks often neglect the characteristics of multiple departments and specializations in modern medical practice. Thirdly, existing evaluation methods are limited to multiple-choice questions, which do not align with the real-world diagnostic scenarios. Lastly, existing evaluation methods lack comprehensive evaluations of end-to-end real clinical scenarios. These limitations in benchmarks in turn obstruct advancements of LLMs and agents for medicine. To address these limitations, we introduce ClinicalLab, a comprehensive clinical diagnosis agent alignment suite. ClinicalLab includes ClinicalBench, an end-to-end multi-departmental clinical diagnostic evaluation benchmark for evaluating medical agents and LLMs. ClinicalBench is based on real cases that cover 24 departments and 150 diseases. ClinicalLab also includes four novel metrics (ClinicalMetrics) for evaluating the effectiveness of LLMs in clinical diagnostic tasks. We evaluate 17 LLMs and find that their performance varies significantly across different departments. Based on these findings, in ClinicalLab, we propose ClinicalAgent, an end-to-end clinical agent that aligns with real-world clinical diagnostic practices. We systematically investigate the performance and applicable scenarios of variants of ClinicalAgent on ClinicalBench. Our findings demonstrate the importance of aligning with modern medical practices in designing medical agents.

The primary goal of continual learning (CL) task in medical image segmentation field is to solve the "catastrophic forgetting" problem, where the model totally forgets previously learned features when it is extended to new categories (class-level) or tasks (task-level). Due to the privacy protection, the historical data labels are inaccessible. Prevalent continual learning methods primarily focus on generating pseudo-labels for old datasets to force the model to memorize the learned features. However, the incorrect pseudo-labels may corrupt the learned feature and lead to a new problem that the better the model is trained on the old task, the poorer the model performs on the new tasks. To avoid this problem, we propose a network by introducing the data-specific Mixture of Experts (MoE) structure to handle the new tasks or categories, ensuring that the network parameters of previous tasks are unaffected or only minimally impacted. To further overcome the tremendous memory costs caused by introducing additional structures, we propose a Low-Rank strategy which significantly reduces memory cost. We validate our method on both class-level and task-level continual learning challenges. Extensive experiments on multiple datasets show our model outperforms all other methods.

The Transformer architecture has significantly advanced deep learning, particularly in natural language processing, by effectively managing long-range dependencies. However, as the demand for understanding complex relationships grows, refining the Transformer's architecture becomes critical. This paper introduces Skip-Layer Attention (SLA) to enhance Transformer models by enabling direct attention between non-adjacent layers. This method improves the model's ability to capture dependencies between high-level abstract features and low-level details. By facilitating direct attention between these diverse feature levels, our approach overcomes the limitations of current Transformers, which often rely on suboptimal intra-layer attention. Our implementation extends the Transformer's functionality by enabling queries in a given layer to interact with keys and values from both the current layer and one preceding layer, thus enhancing the diversity of multi-head attention without additional computational burden. Extensive experiments demonstrate that our enhanced Transformer model achieves superior performance in language modeling tasks, highlighting the effectiveness of our skip-layer attention mechanism., Comment: 7 pages, 1 figure

Training speaker-discriminative and robust speaker verification systems without explicit speaker labels remains a persisting challenge. In this paper, we propose a new self-supervised speaker verification approach, Self-Distillation Prototypes Network (SDPN), which effectively facilitates self-supervised speaker representation learning. SDPN assigns the representation of the augmented views of an utterance to the same prototypes as the representation of the original view, thereby enabling effective knowledge transfer between the views. Originally, due to the lack of negative pairs in the SDPN training process, the network tends to align positive pairs very closely in the embedding space, a phenomenon known as model collapse. To alleviate this problem, we introduce a diversity regularization term to embeddings in SDPN. Comprehensive experiments on the VoxCeleb datasets demonstrate the superiority of SDPN in self-supervised speaker verification. SDPN sets a new state-of-the-art on the VoxCeleb1 speaker verification evaluation benchmark, achieving Equal Error Rate 1.80%, 1.99%, and 3.62% for trial VoxCeleb1-O, VoxCeleb1-E and VoxCeleb1-H respectively, without using any speaker labels in training., Comment: We update this paper to an earlier paper

Sparse triangular solve (SpTRSV) is widely used in various domains. Numerous studies have been conducted using CPUs, GPUs, and specific hardware accelerators, where dataflow can be categorized into coarse and fine granularity. Coarse dataflow offers good spatial locality but suffers from low parallelism, while fine dataflow provides high parallelism but disrupts the spatial structure, leading to increased nodes and poor data reuse. This paper proposes a novel hardware accelerator for SpTRSV or SpTRSV-like DAGs. The accelerator implements a medium granularity dataflow through hardware-software codesign and achieves both excellent spatial locality and high parallelism. Additionally, a partial sum caching mechanism is introduced to reduce the blocking frequency of processing elements (PEs), and a reordering algorithm of intra-node edges computation is developed to enhance data reuse. Experimental results on 264 benchmarks with node counts reaching up to 85,392 demonstrate that this work achieves average performance improvements of 12.2$\times$ (up to 874.5$\times$) over CPUs and 10.1$\times$ (up to 740.4$\times$) over GPUs. Compared to the state-of-the-art technique (DPU-v2), this work shows a 2.5$\times$ (up to 5.9$\times$) average performance improvement and 1.8$\times$ (up to 4.1$\times$) average energy efficiency enhancement.

Low Earth orbit (LEO) satellites have been envisioned as a significant component of the sixth generation (6G) network architecture for achieving ubiquitous coverage and seamless access. However, the implementation of LEO satellites is largely restricted by the deployment of ground stations. Inter-satellite links (ISLs) have been regarded as a promising technique to fully exploit the potentials of LEO mega constellations by concatenating multiple satellites to constitute an autonomous space network. In this article, we present the merits of implementing ISLs in LEO mega constellations and the representative applications empowered/inspired by ISLs. Moreover, we outline several key technical challenges as well as potential solutions related to LEO satellite networks with ISLs, including performance analysis for system design, routing and load balancing, and resource allocation. Particularly, the potential of using ISLs in enhancing in-flight connectivity is showcased with a preliminary performance evaluation. Finally, some open issues are discussed to inspire future research., Comment: 7 pages, 4 figures

Speaker verification systems experience significant performance degradation when tasked with short-duration trial recordings. To address this challenge, a multi-scale feature fusion approach has been proposed to effectively capture speaker characteristics from short utterances. Constrained by the model's size, a robust backbone Enhanced Res2Net (ERes2Net) combining global and local feature fusion demonstrates sub-optimal performance in short-duration speaker verification. To further improve the short-duration feature extraction capability of ERes2Net, we expand the channel dimension within each stage. However, this modification also increases the number of model parameters and computational complexity. To alleviate this problem, we propose an improved ERes2NetV2 by pruning redundant structures, ultimately reducing both the model parameters and its computational cost. A range of experiments conducted on the VoxCeleb datasets exhibits the superiority of ERes2NetV2, which achieves EER of 0.61% for the full-duration trial, 0.98% for the 3s-duration trial, and 1.48% for the 2s-duration trial on VoxCeleb1-O, respectively.

In this paper, we present ControlSpeech, a text-to-speech (TTS) system capable of fully cloning the speaker's voice and enabling arbitrary control and adjustment of speaking style, merely based on a few seconds of audio prompt and a simple textual style description prompt. Prior zero-shot TTS models and controllable TTS models either could only mimic the speaker's voice without further control and adjustment capabilities or were unrelated to speaker-specific voice generation. Therefore, ControlSpeech focuses on a more challenging new task-a TTS system with controllable timbre, content, and style at the same time. ControlSpeech takes speech prompts, content prompts, and style prompts as inputs and utilizes bidirectional attention and mask-based parallel decoding to capture corresponding codec representations in a discrete decoupling codec space. Moreover, we discovered the issue of text style controllability in a many-to-many mapping fashion and proposed the Style Mixture Semantic Density (SMSD) model to resolve this problem. SMSD module which is based on Gaussian mixture density networks, is designed to enhance the fine-grained partitioning and sampling capabilities of style semantic information and generate speech with more diverse styles. In terms of experiments, we make available a controllable model toolkit called ControlToolkit with a new style controllable dataset, some replicated baseline models and propose new metrics to evaluate both the control capability and the quality of generated audio in ControlSpeech. The relevant ablation studies validate the necessity of each component in ControlSpeech is necessary. We hope that ControlSpeech can establish the next foundation paradigm of controllable speech synthesis. The relevant code and demo are available at https://github.com/jishengpeng/ControlSpeech .

Solving large-scale linear programming (LP) problems is an important task in various areas such as communication networks, power systems, finance and logistics. Recently, two distinct approaches have emerged to expedite LP solving: (i) First-order methods (FOMs); (ii) Learning to optimize (L2O). In this work, we propose an FOM-unrolled neural network (NN) called PDHG-Net, and propose a two-stage L2O method to solve large-scale LP problems. The new architecture PDHG-Net is designed by unrolling the recently emerged PDHG method into a neural network, combined with channel-expansion techniques borrowed from graph neural networks. We prove that the proposed PDHG-Net can recover PDHG algorithm, thus can approximate optimal solutions of LP instances with a polynomial number of neurons. We propose a two-stage inference approach: first use PDHG-Net to generate an approximate solution, and then apply PDHG algorithm to further improve the solution. Experiments show that our approach can significantly accelerate LP solving, achieving up to a 3$\times$ speedup compared to FOMs for large-scale LP problems., Comment: Accepted by ICML 2024

Space-air-ground integrated networks (SAGIN) are pivotal for achieving uninterrupted in-flight connectivity (IFC). Most existing studies, however, merely treat satellites as transparent forwarding nodes, and overlook their caching capabilities in enhancing the IFC data rate. In this paper, we consider an IFC-oriented SAGIN, where the satellites collaboratively deliver the content to airborne passengers to facilitate airborne communication. Considering the cached files instantaneously accessible via satellites, this work pioneers the integration of multiple inter-satellite links (ISLs) into the IFC framework, thereby innovating the content delivery process. To minimize the average delay of content delivery, we formulate an optimization problem and propose an exact penalty-based method to derive the satellite association scheme. Our proposed framework has a low complexity and thus paves the way for high-speed Internet connectivity to aviation passengers. Finally, simulation results are presented to demonstrate the effectiveness of our proposed IFC framework for SAGIN., Comment: 6 pages, 4 figures

Existing molecular canonization algorithms typically operate on one-dimensional (1D) string representations or two-dimensional (2D) connectivity graphs of a molecule and are not able to differentiate equivalent atoms based on three-dimensional (3D) structures. The stereochemical tags on each atom are in fact determined according to established Cahn-Ingold-Prelog (CIP) rules for comparing grades, which can help to further differentiate atoms with similar environment. Therefore, a stereochemical-rule-based canonization algorithm that is capable of assigning canonical indices using 3D structural information is of great value. On top of the Schneider-Sayle-Landrum (SSL) partition-based canonization algorithm, we propose an enhanced canonization algorithm to expand its applicability. The initial index assignment rules are redesigned, so that the obtained canonical indices are compatible with the most of the common CIP Sequence Rules, which greatly eases the stereochemical assignment. Furthermore, a branching tiebreaking step is added to secure an accurate evaluation of the structural difference through the minimized root-mean-square deviation (RMSD) between structures, with an option to include hydrogen atoms or not. Our algorithm is implemented with Python and can efficiently obtain minimized RMSD taking into account of the symmetry of molecular systems , contributing to the fields of drug design, molecular docking, and data analysis of molecular dynamics simulation.

Large Language Models (LLMs) have made significant progress in code generation, offering developers groundbreaking automated programming support. However, LLMs often generate code that is syntactically correct and even semantically plausible, but may not execute as expected or fulfill specified requirements. This phenomenon of hallucinations in the code domain has not been systematically explored. To advance the community's understanding and research on this issue, we introduce the concept of code hallucinations and propose a classification method for code hallucination based on execution verification. We categorize code hallucinations into four main types: mapping, naming, resource, and logic hallucinations, with each category further divided into different subcategories to understand and address the unique challenges faced by LLMs in code generation with finer granularity. Additionally, we present a dynamic detection algorithm called CodeHalu designed to detect and quantify code hallucinations. We also introduce the CodeHaluEval benchmark, which includes 8,883 samples from 699 tasks, to systematically and quantitatively evaluate code hallucinations. By evaluating 17 popular LLMs using this benchmark, we reveal significant differences in their accuracy and reliability in code generation, offering detailed insights for further improving the code generation capabilities of LLMs. The CodeHalu benchmark and code are publicly available at https://github.com/yuchen814/CodeHalu.

Using adaptive optics with the Multi-Unit Spectroscopic Explorer (MUSE) on the Very Large Telescope (VLT), the Middle Ages Galaxy Properties with Integral Field Spectroscopy (MAGPI) survey allows us to study the spatially resolved Universe at a crucial time of ~4 Gyr ago ($z$ ~ 0.3) when simulations predict the greatest diversity in evolutionary pathways for galaxies. We investigate the radial trends in the star formation (SF) activity and luminosity-weighted stellar ages as a function of offset from the star-forming main sequence (SFMS) for a total of 294 galaxies. Using both H$\alpha$ emission and the 4000 Angstrom break (i.e., D4000) as star formation rate (SFR) tracers, we find overall flat radial profiles for galaxies lying on and above the SFMS, suggestive of physical processes that enhance/regulate SF throughout the entire galaxy disc. However, for galaxies lying below the SFMS, we find positive gradients in SF suggestive of inside-out quenching. Placing our results in context with results from other redshift regimes suggests an evolution in radial trends at $z$ ~ 0.3 for SF galaxies above the SFMS, from uniformly enhanced SF at $z$ ~ 1 and $z$ ~ 0.3 to centrally enhanced SF at $z$ ~ 0 (when averaged over a wide range of mass). We also capture higher local SFRs for galaxies below the SFMS compared to that of $z$ ~ 0, which can be explained by a larger population of quenched satellites in the local Universe and/or different treatments of limitations set by the D4000-sSFR relation., Comment: 19 pages, 15 figures, 4 tables, accepted for publication in MNRAS

Next-generation mobile networks are expected to facilitate fast AI model downloading to end users. By caching models on edge servers, mobile networks can deliver models to end users with low latency, resulting in a paradigm called edge model caching. In this paper, we develop a novel model placement scheme, called parameter-sharing model caching (TrimCaching). TrimCaching exploits the key observation that a wide range of AI models, such as convolutional neural networks or large language models, can share a significant proportion of parameter blocks containing reusable knowledge, thereby improving storage efficiency. To this end, we formulate a parameter-sharing model placement problem to maximize the cache hit ratio in multi-edge wireless networks by balancing the fundamental tradeoff between storage efficiency and service latency. We show that the formulated problem is a submodular maximization problem with submodular constraints, for which no polynomial-time approximation algorithm exists. To overcome this challenge, we study an important special case, where a small fixed number of parameter blocks are shared across models, which often holds in practice. In such a case, a polynomial-time algorithm with $\left(1-\epsilon\right)/2$-approximation guarantee is developed. Subsequently, we address the original problem for the general case by developing a greedy algorithm. Simulation results demonstrate that the proposed TrimCaching framework significantly improves the cache hit ratio compared with state-of-the-art content caching without exploiting shared parameters in AI models., Comment: 15 pages, 11 figures. Part of this work has been accepted by ICDCS 2024

Muon Induced X-ray Emission (MIXE) was discovered by Chinese physicist Zhang Wenyu as early as 1947, and it can conduct non-destructive elemental analysis inside samples. Research has shown that MIXE can retain the high efficiency of direct imaging while benefiting from the low noise of pinhole imaging through encoding holes. The related technology significantly improves the counting rate while maintaining imaging quality. The sphere encoding technology effectively solves the imaging blurring caused by the tilting of the encoding system, and successfully images micrometer sized X-ray sources. This paper will combine MIXE and X-ray sphere coding imaging techniques, including ball coding and zone plates, to study the method of non-destructive deep structure imaging of ICF targets and obtaining sub element distribution. This method aims to develop a new method for ICF target detection, which is particularly important for inertial confinement fusion. At the same time, this method can be used to detect and analyze materials that are difficult to penetrate or sensitive, and is expected to solve the problem of element resolution and imaging that traditional technologies cannot overcome. It will provide new methods for the future development of multiple fields such as particle physics, material science, and X-ray optics.

We introduce 3D-Speaker-Toolkit, an open-source toolkit for multimodal speaker verification and diarization, designed for meeting the needs of academic researchers and industrial practitioners. The 3D-Speaker-Toolkit adeptly leverages the combined strengths of acoustic, semantic, and visual data, seamlessly fusing these modalities to offer robust speaker recognition capabilities. The acoustic module extracts speaker embeddings from acoustic features, employing both fully-supervised and self-supervised learning approaches. The semantic module leverages advanced language models to comprehend the substance and context of spoken language, thereby augmenting the system's proficiency in distinguishing speakers through linguistic patterns. The visual module applies image processing technologies to scrutinize facial features, which bolsters the precision of speaker diarization in multi-speaker environments. Collectively, these modules empower the 3D-Speaker-Toolkit to achieve substantially improved accuracy and reliability in speaker-related tasks. With 3D-Speaker-Toolkit, we establish a new benchmark for multimodal speaker analysis. The toolkit also includes a handful of open-source state-of-the-art models and a large-scale dataset containing over 10,000 speakers. The toolkit is publicly available at https://github.com/modelscope/3D-Speaker.

The advent of the 6G era aims for ubiquitous connectivity, with the integration of non-terrestrial networks (NTN) offering extensive coverage and enhanced capacity. As manufacturing advances and user demands evolve, space-air-ground integrated networks (SAGIN) with computational capabilities emerge as a viable solution for services requiring low latency and high computational power. Resource management within joint communication and computing-embedded SAGIN (JCC-SAGIN) presents greater complexity than traditional terrestrial networks. This complexity arises from the spatiotemporal dynamics of network topology and service demand, the interdependency of large-scale resource variables, and intricate tradeoffs among various performance metrics. Thus, a thorough examination of resource management strategies in JCC-SAGIN is crucial, emphasizing the role of non-terrestrial platforms with processing capabilities in 6G. This paper begins by reviewing the architecture, enabling technologies, and applications in JCC-SAGIN. Then, we offer a detailed overview of resource management modeling and optimization methods, encompassing both traditional optimization approaches and learning-based intelligent decision-making frameworks. Finally, we outline the prospective research directions in JCC-SAGIN., Comment: 44 pages, 17 figures. This work has been accepted by IEEE Communications Surveys and Tutorials

In this paper, we carry out the entanglement calculations on the coherent intertwiners. We first consider the entanglement introduced by the group-averaging of the tensor-product type intertwiner on a four-valents vertex. The result shows that the entanglement is determined by the probability distribution of recoupling spin, and this probability distribution is a well-behaved peak for the highest (and lowest) weight states. Further, we calculated explicitly the entanglement on gauge-invariant coherent intertwiner with four legs. Our numerical results show that the shape of the semiclassical polyhedron described by the coherent intertwiner can be related to the entanglement; In other words, the entanglement is controlled by the face-angle of the semiclassical polyhedron. Finally, we extend our analytical calculation to the coherent intertwiners with arbitrary number of legs., Comment: 25 pages, 12 figures

The black-box nature of deep learning models in NLP hinders their widespread application. The research focus has shifted to Hierarchical Attribution (HA) for its ability to model feature interactions. Recent works model non-contiguous combinations with a time-costly greedy search in Eculidean spaces, neglecting underlying linguistic information in feature representations. In this work, we introduce a novel method, namely Poincare Explanation (PE), for modeling feature interactions with hyperbolic spaces in a time efficient manner. Specifically, we take building text hierarchies as finding spanning trees in hyperbolic spaces. First we project the embeddings into hyperbolic spaces to elicit inherit semantic and syntax hierarchical structures. Then we propose a simple yet effective strategy to calculate Shapley score. Finally we build the the hierarchy with proving the constructing process in the projected space could be viewed as building a minimum spanning tree and introduce a time efficient building algorithm. Experimental results demonstrate the effectiveness of our approach.

Topologically-engineered mechanical frames are important model constructs for architecture, machine mechanisms, and metamaterials. Despite significant advances in macroscopically fashioned frames, realization and phonon imaging of nanoframes have remained challenging. Here we extend for the first time the principles of topologically-engineered mechanical frames to lattices self-assembled from nanoparticles. Liquid-phase transmission electron microscopy images the vibrations of nanoparticles in self-assembled Maxwell and hexagonal lattices at the nanometer resolution, measuring a series of otherwise inaccessible properties such as phonon spectra and nonlinear lattice deformation paths. These properties are experimentally modulated by ionic strength, captured by our discrete mechanical model considering the complexity of nanoscale interactions and thermal fluctuations. The experiment-theory integration bridges mechanical metamaterials and colloidal self-assembly, opening new opportunities to manufacture phononic devices with solution processibility, transformability, light weight, and emergent functions, at underexplored length, frequency, and energy scales.

To meet the demands for ubiquitous communication and temporary edge computing in 6G networks, aerial mobile edge computing (MEC) networks have been envisioned as a new paradigm. However, dynamic user requests pose challenges for task assignment strategies. Most of the existing research assumes that the strategy is deployed on ground-based stations or UAVs, which will be ineffective in an environment lacking infrastructure and continuous energy supply. Moreover, the resource mutual exclusion problem of dynamic task assignment has not been effectively solved. Toward this end, we introduce the digital twin (DT) into the aerial MEC network to study the resource coalition cooperation approach with the generative model (GM), which provides a preliminary coalition structure for the coalition game. Specifically, we propose a novel network framework that is composed of an application plane, a physical plane, and a virtual plane. After that, the task assignment problem is simplified to convex optimization programming with linear constraints. And then, we also propose a resource coalition cooperation approach that is based on a transferable utility (TU) coalition game to obtain an approximate optimal solution. Numerical results confirm the effectiveness of our proposed approach in terms of energy consumption and utilization of resources.

The scatter about the mass-metallicity relation (MZR) has a correlation with the star formation rate (SFR) of galaxies. The lack of evidence of evolution in correlated scatter at $z\lesssim2.5$ leads many to refer to the relationship between mass, metallicity, and SFR as the Fundamental Metallicity Relation (FMR). Yet, recent high-redshift (z>3) JWST observations have challenged the fundamental (i.e., redshift-invariant) nature of the FMR. In this work, we show that the cosmological simulations Illustris, IllustrisTNG, and EAGLE all predict MZRs that exhibit scatter with a secondary dependence on SFR up to $z=8$. We introduce the concept of a "strong" FMR, where the strength of correlated scatter does not evolve with time, and a "weak" FMR, where there is some time evolution. We find that each simulation analysed has a weak FMR -- there is non-negligible evolution in the strength of the correlation with SFR. Furthermore, we show that the scatter is reduced an additional ~10-40% at $z\gtrsim3$ when using a weak FMR, compared to assuming a strong FMR. These results highlight the importance of avoiding coarse redshift binning when assessing the FMR., Comment: 10 pages, 5 figures (Figure 5 is summary). Accepted to MNRAS

Neural networks, with powerful nonlinear mapping and classification capabilities, are widely applied in mechanical fault diagnosis to ensure safety. However, being typical black-box models, their application is limited in high-reliability-required scenarios. To understand the classification logic and explain what typical fault signals look like, the prototype matching network (PMN) is proposed by combining the human-inherent prototype-matching with autoencoder (AE). The PMN matches AE-extracted feature with each prototype and selects the most similar prototype as the prediction result. It has three interpreting paths on classification logic, fault prototypes, and matching contributions. Conventional diagnosis and domain generalization experiments demonstrate its competitive diagnostic performance and distinguished advantages in representation learning. Besides, the learned typical fault signals (i.e., sample-level prototypes) showcase the ability for denoising and extracting subtle key features that experts find challenging to capture. This ability broadens human understanding and provides a promising solution from interpretability research to AI-for-Science., Comment: 17 pages, 12 figures, 6 tables

Reinforcement Learning (RL)-Based Recommender Systems (RSs) have gained rising attention for their potential to enhance long-term user engagement. However, research in this field faces challenges, including the lack of user-friendly frameworks, inconsistent evaluation metrics, and difficulties in reproducing existing studies. To tackle these issues, we introduce EasyRL4Rec, an easy-to-use code library designed specifically for RL-based RSs. This library provides lightweight and diverse RL environments based on five public datasets and includes core modules with rich options, simplifying model development. It provides unified evaluation standards focusing on long-term outcomes and offers tailored designs for state modeling and action representation for recommendation scenarios. Furthermore, we share our findings from insightful experiments with current methods. EasyRL4Rec seeks to facilitate the model development and experimental process in the domain of RL-based RSs. The library is available for public use., Comment: Accepted by SIGIR2024

Fluctuations of conserved charges, such as the net-baryon number fluctuations, are influenced by different dynamical evolution processes. In this paper, we investigate the influence of hadronic rescatterings on different orders of cumulants of the net-baryon number distribution. At the start of hadronic rescatterings, we introduce net-baryon number distributions reconstructed based on net-baryon cumulants of different orders obtained from computation in functional renormalization group (FRG), where the distributions were constructed using the maximum entropy method. This way we introduce the critical fluctuations of Quantum Chromodynamics (QCD) into the AMPT model. Firstly, we find that hadronic rescatterings have distinct effects on cumulant ratios of different orders for the net-baryon number. Secondly, we observe that the effect of hadronic rescatterings is more significant for critical fluctuations than dynamical fluctuations, because the two-, three- and four-particle correlation functions due to critical fluctuations are weakened more significantly by hadronic rescatterings., Comment: 10 pages, 8 figures

In recent years, large language models have achieved significant success in generative tasks (e.g., speech cloning and audio generation) related to speech, audio, music, and other signal domains. A crucial element of these models is the discrete acoustic codecs, which serves as an intermediate representation replacing the mel-spectrogram. However, there exist several gaps between discrete codecs and downstream speech language models. Specifically, 1) most codec models are trained on only 1,000 hours of data, whereas most speech language models are trained on 60,000 hours; 2) Achieving good reconstruction performance requires the utilization of numerous codebooks, which increases the burden on downstream speech language models; 3) The initial channel of the codebooks contains excessive information, making it challenging to directly generate acoustic tokens from weakly supervised signals such as text in downstream tasks. Consequently, leveraging the characteristics of speech language models, we propose Language-Codec. In the Language-Codec, we introduce a Mask Channel Residual Vector Quantization (MCRVQ) mechanism along with improved Fourier transform structures and larger training datasets to address the aforementioned gaps. We compare our method with competing audio compression algorithms and observe significant outperformance across extensive evaluations. Furthermore, we also validate the efficiency of the Language-Codec on downstream speech language models. The source code and pre-trained models can be accessed at https://github.com/jishengpeng/languagecodec ., Comment: We release a more powerful checkpoint in Language-Codec v3

In this paper, we focus on solving one of the most important tasks in the field of speech processing, i.e., automatic speech recognition (ASR), with speech foundation encoders and large language models (LLM). Recent works have complex designs such as compressing the output temporally for the speech encoder, tackling modal alignment for the projector, and utilizing parameter-efficient fine-tuning for the LLM. We found that delicate designs are not necessary, while an embarrassingly simple composition of off-the-shelf speech encoder, LLM, and the only trainable linear projector is competent for the ASR task. To be more specific, we benchmark and explore various combinations of LLMs and speech encoders, leading to the optimal LLM-based ASR system, which we call SLAM-ASR. The proposed SLAM-ASR provides a clean setup and little task-specific design, where only the linear projector is trained. To the best of our knowledge, SLAM-ASR achieves the best performance on the Librispeech benchmark among LLM-based ASR models and even outperforms the latest LLM-based audio-universal model trained on massive pair data. Finally, we explore the capability emergence of LLM-based ASR in the process of modal alignment. We hope that our study can facilitate the research on extending LLM with cross-modality capacity and shed light on the LLM-based ASR community., Comment: Working in progress and will open-source soon

Graph analysis is fundamental in real-world applications. Traditional approaches rely on SPARQL-like languages or clicking-and-dragging interfaces to interact with graph data. However, these methods either require users to possess high programming skills or support only a limited range of graph analysis functionalities. To address the limitations, we propose a large language model (LLM)-based framework called ChatGraph. With ChatGraph, users can interact with graphs through natural language, making it easier to use and more flexible than traditional approaches. The core of ChatGraph lies in generating chains of graph analysis APIs based on the understanding of the texts and graphs inputted in the user prompts. To achieve this, ChatGraph consists of three main modules: an API retrieval module that searches for relevant APIs, a graph-aware LLM module that enables the LLM to comprehend graphs, and an API chain-oriented finetuning module that guides the LLM in generating API chains.