Search

Simulating strongly-correlated quantum many-body systems at finite temperatures is a significant challenge in computational physics. In this work, we present a scalable finite-temperature tensor network algorithm for two-dimensional quantum many-body systems. We employ the (fermionic) projected entangled pair state (PEPS) to represent the vectorization of the quantum thermal state and utilize a stochastic reconfiguration method to cool down the quantum states from infinite temperature. We validate our method by benchmarking it against the 2D antiferromagnetic Heisenberg model, the $J_1$-$J_2$ model, and the Fermi-Hubbard model, comparing physical properties such as internal energy, specific heat, and magnetic susceptibility with results obtained from stochastic series expansion (SSE), exact diagonalization, and determinant quantum Monte Carlo (DQMC).

LLM-powered personalization agent systems employ Large Language Models (LLMs) to predict users' behavior from their past activities. However, their effectiveness often hinges on the ability to effectively leverage extensive, long user historical data due to its inherent noise and length of such data. Existing pretrained LLMs may generate summaries that are concise but lack the necessary context for downstream tasks, hindering their utility in personalization systems. To address these challenges, we introduce Reinforcement Learning from Prediction Feedback (RLPF). RLPF fine-tunes LLMs to generate concise, human-readable user summaries that are optimized for downstream task performance. By maximizing the usefulness of the generated summaries, RLPF effectively distills extensive user history data while preserving essential information for downstream tasks. Our empirical evaluation demonstrates significant improvements in both extrinsic downstream task utility and intrinsic summary quality, surpassing baseline methods by up to 22% on downstream task performance and achieving an up to 84.59% win rate on Factuality, Abstractiveness, and Readability. RLPF also achieves a remarkable 74% reduction in context length while improving performance on 16 out of 19 unseen tasks and/or datasets, showcasing its generalizability. This approach offers a promising solution for enhancing LLM personalization by effectively transforming long, noisy user histories into informative and human-readable representations.

Fully Homomorphic Encryption (FHE) enables computations on encrypted data, preserving confidentiality without the need for decryption. However, FHE is often hindered by significant performance overhead, particularly for high-precision and complex data like images. Due to serious efficiency issues, traditional FHE methods often encrypt images by monolithic data blocks (such as pixel rows), instead of pixels. However, this strategy compromises the advantages of homomorphic operations and disables pixel-level image processing. In this study, we address these challenges by proposing and implementing a pixel-level homomorphic encryption approach, iCHEETAH, based on the CKKS scheme. To enhance computational efficiency, we introduce three novel caching mechanisms to pre-encrypt radix values or frequently occurring pixel values, substantially reducing redundant encryption operations. Extensive experiments demonstrate that our approach achieves up to a 19-fold improvement in encryption speed compared to the original CKKS, while maintaining high image quality. Additionally, real-world image applications such as mean filtering, brightness enhancement, image matching and watermarking are tested based on FHE, showcasing up to a 91.53% speed improvement. We also proved that our method is IND-CPA (Indistinguishability under Chosen Plaintext Attack) secure, providing strong encryption security. These results underscore the practicality and efficiency of iCHEETAH, marking a significant advancement in privacy-preserving image processing at scale., Comment: 16 pages, 10 figures

We propose a method for formally certifying and quantifying individual fairness of deep neural networks (DNN). Individual fairness guarantees that any two individuals who are identical except for a legally protected attribute (e.g., gender or race) receive the same treatment. While there are existing techniques that provide such a guarantee, they tend to suffer from lack of scalability or accuracy as the size and input dimension of the DNN increase. Our method overcomes this limitation by applying abstraction to a symbolic interval based analysis of the DNN followed by iterative refinement guided by the fairness property. Furthermore, our method lifts the symbolic interval based analysis from conventional qualitative certification to quantitative certification, by computing the percentage of individuals whose classification outputs are provably fair, instead of merely deciding if the DNN is fair. We have implemented our method and evaluated it on deep neural networks trained on four popular fairness research datasets. The experimental results show that our method is not only more accurate than state-of-the-art techniques but also several orders-of-magnitude faster., Comment: To Appear In Proceedings of the 47th IEEE/ACM International Conference on Software Engineering (ICSE 2025)

Large language models (LLMs) have shown remarkable capabilities in generating user summaries from a long list of raw user activity data. These summaries capture essential user information such as preferences and interests, and therefore are invaluable for LLM-based personalization applications, such as explainable recommender systems. However, the development of new summarization techniques is hindered by the lack of ground-truth labels, the inherent subjectivity of user summaries, and human evaluation which is often costly and time-consuming. To address these challenges, we introduce \UserSumBench, a benchmark framework designed to facilitate iterative development of LLM-based summarization approaches. This framework offers two key components: (1) A reference-free summary quality metric. We show that this metric is effective and aligned with human preferences across three diverse datasets (MovieLens, Yelp and Amazon Review). (2) A novel robust summarization method that leverages time-hierarchical summarizer and self-critique verifier to produce high-quality summaries while eliminating hallucination. This method serves as a strong baseline for further innovation in summarization techniques.

A text-to-speech (TTS) model trained to reconstruct speech given text tends towards predictions that are close to the average characteristics of a dataset, failing to model the variations that make human speech sound natural. This problem is magnified for zero-shot voice cloning, a task that requires training data with high variance in speaking styles. We build off of recent works which have used Generative Advsarial Networks (GAN) by proposing a Transformer encoder-decoder architecture to conditionally discriminates between real and generated speech features. The discriminator is used in a training pipeline that improves both the acoustic and prosodic features of a TTS model. We introduce our novel adversarial training technique by applying it to a FastSpeech2 acoustic model and training on Libriheavy, a large multi-speaker dataset, for the task of zero-shot voice cloning. Our model achieves improvements over the baseline in terms of speech quality and speaker similarity. Audio examples from our system are available online., Comment: Accepted at INTERSPEECH 2024

A primary challenge in wireless power transfer (WPT) systems is to achieve efficient and stable power transmission without complex control strategies when load conditions change dynamically. Addressing this issue, we propose a third-order pseudo-Hermitian WPT system whose output characteristics exhibit a stable frequency and constant power. The frequency selection mechanism and energy efficiency of the nonlinear WPT system based on pseudo-Hermitian under the coupling mode theory approximation are analyzed. Theoretical analysis indicates that under certain coupling coefficients and load conditions, the proposed system can achieve frequency adaptation in a stable frequency mode without the need to change the circuit frequency. When the load changes dynamically, the stability of the power output is maintained using a proportional integral (PI) control strategy that only collects the voltage and current at the transmitting end, eliminating the need for wireless communication circuits with feedback from the receiving side. Experimental results demonstrate that the proposed design scheme can achieve constant power transmission when load conditions change, maintaining stable and relatively high transmission efficiency. The proposed scheme exhibits benefits in practical applications since no communication is required.

Known as low energy consumption networks, spiking neural networks (SNNs) have gained a lot of attention within the past decades. While SNNs are increasing competitive with artificial neural networks (ANNs) for vision tasks, they are rarely used for long sequence tasks, despite their intrinsic temporal dynamics. In this work, we develop spiking state space models (SpikingSSMs) for long sequence learning by leveraging on the sequence learning abilities of state space models (SSMs). Inspired by dendritic neuron structure, we hierarchically integrate neuronal dynamics with the original SSM block, meanwhile realizing sparse synaptic computation. Furthermore, to solve the conflict of event-driven neuronal dynamics with parallel computing, we propose a light-weight surrogate dynamic network which accurately predicts the after-reset membrane potential and compatible to learnable thresholds, enabling orders of acceleration in training speed compared with conventional iterative methods. On the long range arena benchmark task, SpikingSSM achieves competitive performance to state-of-the-art SSMs meanwhile realizing on average 90\% of network sparsity. On language modeling, our network significantly surpasses existing spiking large language models (spikingLLMs) on the WikiText-103 dataset with only a third of the model size, demonstrating its potential as backbone architecture for low computation cost LLMs.

It is known that any periodic map of order $n$ on a closed oriented surface of genus $g$ can be equivariantly embedded into $S^m$ for some $m$. In the orientable and smooth category, we determine the smallest possible $m$ when $n\geq 3g$. We show that for each integer $k>1$ there exist infinitely many periodic maps such that the smallest possible $m$ is equal to $k$., Comment: 20 pages, 8 Figures

A long memory and non-linear realized volatility model class is proposed for direct Value at Risk (VaR) forecasting. This model, referred to as RNN-HAR, extends the heterogeneous autoregressive (HAR) model, a framework known for efficiently capturing long memory in realized measures, by integrating a Recurrent Neural Network (RNN) to handle non-linear dynamics. Loss-based generalized Bayesian inference with Sequential Monte Carlo is employed for model estimation and sequential prediction in RNN HAR. The empirical analysis is conducted using daily closing prices and realized measures from 2000 to 2022 across 31 market indices. The proposed models one step ahead VaR forecasting performance is compared against a basic HAR model and its extensions. The results demonstrate that the proposed RNN-HAR model consistently outperforms all other models considered in the study.

Dynamic activation (DA) techniques, such as DejaVu and MoEfication, have demonstrated their potential to significantly enhance the inference efficiency of large language models (LLMs). However, these techniques often rely on ReLU activation functions or require additional parameters and training to maintain performance. This paper introduces a training-free Threshold-based Dynamic Activation(TDA) method that leverage sequence information to exploit the inherent sparsity of models across various architectures. This method is designed to accelerate generation speed by 18-25\% without significantly compromising task performance, thereby addressing the limitations of existing DA techniques. Moreover, we delve into the root causes of LLM sparsity and theoretically analyze two of its critical features: history-related activation uncertainty and semantic-irrelevant activation inertia. Our comprehensive analyses not only provide a robust theoretical foundation for DA methods but also offer valuable insights to guide future research in optimizing LLMs for greater efficiency and effectiveness.

A quantum network provides an infrastructure connecting quantum devices with revolutionary computing, sensing, and communication capabilities. As the best-known application of a quantum network, quantum key distribution (QKD) shares secure keys guaranteed by the laws of quantum mechanics. A quantum satellite constellation offers a solution to facilitate the quantum network on a global scale. The Micius satellite has verified the feasibility of satellite quantum communications, however, scaling up quantum satellite constellations is challenging, requiring small lightweight satellites, portable ground stations and real-time secure key exchange. Here we tackle these challenges and report the development of a quantum microsatellite capable of performing space-to-ground QKD using portable ground stations. The quantum microsatellite features a payload weighing approximately 23 kg, while the portable ground station weighs about 100 kg. These weights represent reductions by more than an order and two orders of magnitude, respectively, compared to the Micius satellite. Additionally, we multiplex bidirectional satellite-ground optical communication with quantum communication, enabling key distillation and secure communication in real-time. Using the microsatellite and the portable ground stations, we demonstrate satellite-based QKD with multiple ground stations and achieve the sharing of up to 0.59 million bits of secure keys during a single satellite pass. The compact quantum payload can be readily assembled on existing space stations or small satellites, paving the way for a satellite-constellation-based quantum and classical network for widespread real-life applications., Comment: 40 pages, 8 figures

Event camera-based visual tracking has drawn more and more attention in recent years due to the unique imaging principle and advantages of low energy consumption, high dynamic range, and dense temporal resolution. Current event-based tracking algorithms are gradually hitting their performance bottlenecks, due to the utilization of vision Transformer and the static template for target object localization. In this paper, we propose a novel Mamba-based visual tracking framework that adopts the state space model with linear complexity as a backbone network. The search regions and target template are fed into the vision Mamba network for simultaneous feature extraction and interaction. The output tokens of search regions will be fed into the tracking head for target localization. More importantly, we consider introducing a dynamic template update strategy into the tracking framework using the Memory Mamba network. By considering the diversity of samples in the target template library and making appropriate adjustments to the template memory module, a more effective dynamic template can be integrated. The effective combination of dynamic and static templates allows our Mamba-based tracking algorithm to achieve a good balance between accuracy and computational cost on multiple large-scale datasets, including EventVOT, VisEvent, and FE240hz. The source code will be released on https://github.com/Event-AHU/MambaEVT, Comment: In Peer Review

Recent years have witnessed substantial advancements in the field of 3D reconstruction from 2D images, particularly following the introduction of the neural radiance field (NeRF) technique. However, reconstructing a 3D high dynamic range (HDR) radiance field, which aligns more closely with real-world conditions, from 2D multi-exposure low dynamic range (LDR) images continues to pose significant challenges. Approaches to this issue fall into two categories: grid-based and implicit-based. Implicit methods, using multi-layer perceptrons (MLP), face inefficiencies, limited solvability, and overfitting risks. Conversely, grid-based methods require significant memory and struggle with image quality and long training times. In this paper, we introduce Gaussian Splatting-a recent, high-quality, real-time 3D reconstruction technique-into this domain. We further develop the High Dynamic Range Gaussian Splatting (HDR-GS) method, designed to address the aforementioned challenges. This method enhances color dimensionality by including luminance and uses an asymmetric grid for tone-mapping, swiftly and precisely converting pixel irradiance to color. Our approach improves HDR scene recovery accuracy and integrates a novel coarse-to-fine strategy to speed up model convergence, enhancing robustness against sparse viewpoints and exposure extremes, and preventing local optima. Extensive testing confirms that our method surpasses current state-of-the-art techniques in both synthetic and real-world scenarios. Code will be released at \url{https://github.com/WuJH2001/HDRGS}

The thriving mobile app ecosystem encompasses a wide range of functionalities. However, within this ecosystem, a subset of apps provides illicit services such as gambling and pornography to pursue economic gains, collectively referred to as "underground economy apps". While previous studies have examined these apps' characteristics and identification methods, investigations into their distribution via platforms beyond app markets (like Telegram) remain scarce, which has emerged as a crucial channel for underground activities and cybercrime due to the robust encryption and user anonymity. This study provides the first comprehensive exploration of the underground mobile app ecosystem on Telegram. Overcoming the complexities of the Telegram environment, we build a novel dataset and analyze the prevalence, promotional strategies, and characteristics of these apps. Our findings reveal that these apps reach approximately 1% of Telegram's user base, primarily catering to gambling and pornography services. We uncover sophisticated promotional strategies involving complex networks of apps, websites, users, and channels, and identify significant gaps in Telegram's content moderation capabilities. Our analysis also exposes the misuse of iOS features for app distribution and the prevalence of malicious behaviors in these apps. This research not only enhances our understanding of the underground app ecosystem but also provides valuable insights for developing effective regulatory measures and protecting users from potential risks associated with these covert operations. Our findings provide implications for platform regulators, app market operators, law enforcement agencies, and cybersecurity professionals in combating the proliferation of underground apps on encrypted messaging platforms.

Projected Entangled Pair States (PEPS) are recognized as a potent tool for exploring two-dimensional quantum many-body systems. However, a significant challenge emerges when applying conventional PEPS methodologies to systems with periodic boundary conditions (PBC), attributed to the prohibitive computational scaling with the bond dimension. This has notably restricted the study of systems with complex boundary conditions. To address this challenge, we have developed a strategy that involves the superposition of PEPS with open boundary conditions (OBC) to treat systems with PBC. This approach significantly reduces the computational complexity of such systems while maintaining their translational invariance and the PBC. We benchmark this method against the Heisenberg model and the $J_1$-$J_2$ model, demonstrating its capability to yield highly accurate results at low computational costs, even for large system sizes. The techniques are adaptable to other boundary conditions, including cylindrical and twisted boundary conditions, and therefore significantly expands the application scope of the PEPS approach, shining new light on numerous applications.

Let $c(K)$ denote the crossing number of a knot $K$ and let $K_1\# K_2$ denote the connected sum of two oriented knots $K_1$ and $K_2$. It is a very old unsolved question that whether $c(K_1\# K_2)=c(K_1)+c(K_2)$. In this paper we show that $c(K_1\# K_2)> (c(K_1)+c(K_2))/16$., Comment: 46 pages, 45 figures

The advent of the sixth-generation (6G) networks presents another round of revolution for the mobile communication landscape, promising an immersive experience, robust reliability, minimal latency, extreme connectivity, ubiquitous coverage, and capabilities beyond communication, including intelligence and sensing. To achieve these ambitious goals, it is apparent that 6G networks need to incorporate the state-of-the-art technologies. One of the technologies that has garnered rising interest is fluid antenna system (FAS) which represents any software-controllable fluidic, conductive, or dielectric structure capable of dynamically changing its shape and position to reconfigure essential radio-frequency (RF) characteristics. Compared to traditional antenna systems (TASs) with fixed-position radiating elements, the core idea of FAS revolves around the unique flexibility of reconfiguring the radiating elements within a given space. One recent driver of FAS is the recognition of its position-flexibility as a new degree of freedom (dof) to harness diversity and multiplexing gains. In this paper, we provide a comprehensive tutorial, covering channel modeling, signal processing and estimation methods, information-theoretic insights, new multiple access techniques, and hardware designs. Moreover, we delineate the challenges of FAS and explore the potential of using FAS to improve the performance of other contemporary technologies. By providing insights and guidance, this tutorial paper serves to inspire researchers to explore new horizons and fully unleash the potential of FAS., Comment: 50 pages, 45 figures, 5 tables. Submitted for potential publication

Large Language Models (LLMs) represent a significant stride toward Artificial General Intelligence. As scaling laws underscore the potential of increasing model sizes, the academic community has intensified its investigations into LLMs with capacities exceeding 50 billion parameters. This technical report builds on our prior work with Tele-FLM (also known as FLM-2), a publicly available 52-billion-parameter model. We delve into two primary areas: we first discuss our observation of Supervised Fine-tuning (SFT) on Tele-FLM-52B, which supports the "less is more" approach for SFT data construction; second, we demonstrate our experiments and analyses on the best practices for progressively growing a model from 52 billion to 102 billion, and subsequently to 1 trillion parameters. We will open-source a 1T model checkpoint, namely Tele-FLM-1T, to advance further training and research., Comment: For the Tele-FLM-52B tech report, see also 2404.16645

Prognosis and Health Management (PHM), critical for ensuring task completion by complex systems and preventing unexpected failures, is widely adopted in aerospace, manufacturing, maritime, rail, energy, etc. However, PHM's development is constrained by bottlenecks like generalization, interpretation and verification abilities. Presently, generative artificial intelligence (AI), represented by Large Model, heralds a technological revolution with the potential to fundamentally reshape traditional technological fields and human production methods. Its capabilities, including strong generalization, reasoning, and generative attributes, present opportunities to address PHM's bottlenecks. To this end, based on a systematic analysis of the current challenges and bottlenecks in PHM, as well as the research status and advantages of Large Model, we propose a novel concept and three progressive paradigms of Prognosis and Health Management Large Model (PHM-LM) through the integration of the Large Model with PHM. Subsequently, we provide feasible technical approaches for PHM-LM to bolster PHM's core capabilities within the framework of the three paradigms. Moreover, to address core issues confronting PHM, we discuss a series of technical challenges of PHM-LM throughout the entire process of construction and application. This comprehensive effort offers a holistic PHM-LM technical framework, and provides avenues for new PHM technologies, methodologies, tools, platforms and applications, which also potentially innovates design, research & development, verification and application mode of PHM. And furthermore, a new generation of PHM with AI will also capably be realized, i.e., from custom to generalized, from discriminative to generative, and from theoretical conditions to practical applications.

Graph Neural Networks (GNNs) have demonstrated significant application potential in various fields. However, GNNs are still vulnerable to adversarial attacks. Numerous adversarial defense methods on GNNs are proposed to address the problem of adversarial attacks. However, these methods can only serve as a defense before poisoning, but cannot repair poisoned GNN. Therefore, there is an urgent need for a method to repair poisoned GNN. In this paper, we address this gap by introducing the novel concept of model repair for GNNs. We propose a repair framework, Repairing Robustness of Graph Neural Networks via Machine Unlearning (GraphMU), which aims to fine-tune poisoned GNN to forget adversarial samples without the need for complete retraining. We also introduce a unlearning validation method to ensure that our approach effectively forget specified poisoned data. To evaluate the effectiveness of GraphMU, we explore three fine-tuned subgraph construction scenarios based on the available perturbation information: (i) Known Perturbation Ratios, (ii) Known Complete Knowledge of Perturbations, and (iii) Unknown any Knowledge of Perturbations. Our extensive experiments, conducted across four citation datasets and four adversarial attack scenarios, demonstrate that GraphMU can effectively restore the performance of poisoned GNN.

Graph neural networks (GNNs) have achieved tremendous success, but recent studies have shown that GNNs are vulnerable to adversarial attacks, which significantly hinders their use in safety-critical scenarios. Therefore, the design of robust GNNs has attracted increasing attention. However, existing research has mainly been conducted via experimental trial and error, and thus far, there remains a lack of a comprehensive understanding of the vulnerability of GNNs. To address this limitation, we systematically investigate the adversarial robustness of GNNs by considering graph data patterns, model-specific factors, and the transferability of adversarial examples. Through extensive experiments, a set of principled guidelines is obtained for improving the adversarial robustness of GNNs, for example: (i) rather than highly regular graphs, the training graph data with diverse structural patterns is crucial for model robustness, which is consistent with the concept of adversarial training; (ii) the large model capacity of GNNs with sufficient training data has a positive effect on model robustness, and only a small percentage of neurons in GNNs are affected by adversarial attacks; (iii) adversarial transfer is not symmetric and the adversarial examples produced by the small-capacity model have stronger adversarial transferability. This work illuminates the vulnerabilities of GNNs and opens many promising avenues for designing robust GNNs.

Massive Over-activation Yielded Uplifts(MOYU) is an inherent property of large language models, and dynamic activation(DA) based on the MOYU property is a clever yet under-explored strategy designed to accelerate inference in these models. Existing methods that utilize MOYU often face a significant 'Impossible Trinity': struggling to simultaneously maintain model performance, enhance inference speed, and extend applicability across various architectures. Due to the theoretical ambiguities surrounding MOYU, this paper elucidates the root cause of the MOYU property and outlines the mechanisms behind two primary limitations encountered by current DA methods: 1) history-related activation uncertainty, and 2) semantic-irrelevant activation inertia. Our analysis not only underscores the limitations of current dynamic activation strategies within large-scale LLaMA models but also proposes opportunities for refining the design of future sparsity schemes.

Spiking neural networks (SNNs) are gaining increasing attention as potential computationally efficient alternatives to traditional artificial neural networks(ANNs). However, the unique information propagation mechanisms and the complexity of SNN neuron models pose challenges for adopting traditional methods developed for ANNs to SNNs. These challenges include both weight learning and architecture design. While surrogate gradient learning has shown some success in addressing the former challenge, the latter remains relatively unexplored. Recently, a novel paradigm utilizing evolutionary computation methods has emerged to tackle these challenges. This approach has resulted in the development of a variety of energy-efficient and high-performance SNNs across a wide range of machine learning benchmarks. In this paper, we present a survey of these works and initiate discussions on potential challenges ahead.

In a rapidly evolving job market, skill demand forecasting is crucial as it enables policymakers and businesses to anticipate and adapt to changes, ensuring that workforce skills align with market needs, thereby enhancing productivity and competitiveness. Additionally, by identifying emerging skill requirements, it directs individuals towards relevant training and education opportunities, promoting continuous self-learning and development. However, the absence of comprehensive datasets presents a significant challenge, impeding research and the advancement of this field. To bridge this gap, we present Job-SDF, a dataset designed to train and benchmark job-skill demand forecasting models. Based on 10.35 million public job advertisements collected from major online recruitment platforms in China between 2021 and 2023, this dataset encompasses monthly recruitment demand for 2,324 types of skills across 521 companies. Our dataset uniquely enables evaluating skill demand forecasting models at various granularities, including occupation, company, and regional levels. We benchmark a range of models on this dataset, evaluating their performance in standard scenarios, in predictions focused on lower value ranges, and in the presence of structural breaks, providing new insights for further research. Our code and dataset are publicly accessible via the https://github.com/Job-SDF/benchmark.

Quantization is a proven effective method for compressing large language models. Although popular techniques like W8A8 and W4A16 effectively maintain model performance, they often fail to concurrently speed up the prefill and decoding stages of inference. W4A8 is a promising strategy to accelerate both of them while usually leads to a significant performance degradation. To address these issues, we present QQQ, a Quality Quattuor-bit Quantization method with 4-bit weights and 8-bit activations. QQQ employs adaptive smoothing and Hessian-based compensation, significantly enhancing the performance of quantized models without extensive training. Furthermore, we meticulously engineer W4A8 GEMM kernels to increase inference speed. Our specialized per-channel W4A8 GEMM and per-group W4A8 GEMM achieve impressive speed increases of 3.67$\times$ and 3.29 $\times$ over FP16 GEMM. Our extensive experiments show that QQQ achieves performance on par with existing state-of-the-art LLM quantization methods while significantly accelerating inference, achieving speed boosts up to 2.24 $\times$, 2.10$\times$, and 1.25$\times$ compared to FP16, W8A8, and W4A16, respectively.

The Android ecosystem faces a notable challenge known as fragmentation, which denotes the extensive diversity within the system. This issue is mainly related to differences in system versions, device hardware specifications, and customizations introduced by manufacturers. The growing divergence among devices leads to marked variations in how a given app behaves across diverse devices. This is referred to as device-specific behaviors. In this work, we present the first large-scale empirical study of device-specific behaviors in real-world Android apps. We have designed a three-phase static analysis framework to accurately detect and understand the device-specific behaviors. Upon employing our tool on a dataset comprising more than 20,000 apps, we detected device-specific behaviors in 2,357 of them. By examining the distribution of device-specific behaviors, our analysis revealed that apps within the Chinese third-party app market exhibit more relevant behaviors compared to their counterparts in Google Play. Additionally, these behaviors are more likely to feature dominant brands that hold larger market shares. Reflecting this, we have classified these device-specific behaviors into 29 categories based on implemented functionalities, providing structured insight into these behaviors. Beyond common behaviors like issue fixes and feature adaptations, we observed 33 aggressive apps, including popular ones with millions of downloads, abusing system properties of customized ROMs to obtain user-unresettable identifiers without requiring permission, substantially impacting user privacy. Finally, we investigated the origins of device-specific behaviors, revealing significant challenges developers face in implementing them comprehensively. Our research sheds light on the promising but less touched research direction of device-specific behaviors, benefiting community stakeholders.

Exploring proper way to conduct multi-speech feature fusion for cross-corpus speech emotion recognition is crucial as different speech features could provide complementary cues reflecting human emotion status. While most previous approaches only extract a single speech feature for emotion recognition, existing fusion methods such as concatenation, parallel connection, and splicing ignore heterogeneous patterns in the interaction between features and features, resulting in performance of existing systems. In this paper, we propose a novel graph-based fusion method to explicitly model the relationships between every pair of speech features. Specifically, we propose a multi-dimensional edge features learning strategy called Graph-based multi-Feature fusion method for speech emotion recognition. It represents each speech feature as a node and learns multi-dimensional edge features to explicitly describe the relationship between each feature-feature pair in the context of emotion recognition. This way, the learned multi-dimensional edge features encode speech feature-level information from both the vertex and edge dimensions. Our Approach consists of three modules: an Audio Feature Generation(AFG)module, an Audio-Feature Multi-dimensional Edge Feature(AMEF) module and a Speech Emotion Recognition (SER) module. The proposed methodology yielded satisfactory outcomes on the SEWA dataset. Furthermore, the method demonstrated enhanced performance compared to the baseline in the AVEC 2019 Workshop and Challenge. We used data from two cultures as our training and validation sets: two cultures containing German and Hungarian on the SEWA dataset, the CCC scores for German are improved by 17.28% for arousal and 7.93% for liking. The outcomes of our methodology demonstrate a 13% improvement over alternative fusion techniques, including those employing one dimensional edge-based feature fusion approach., Comment: 25 pages,4 figures

Radiance field methods represent the state of the art in reconstructing complex scenes from multi-view photos. However, these reconstructions often suffer from one or both of the following limitations: First, they typically represent scenes in low dynamic range (LDR), which restricts their use to evenly lit environments and hinders immersive viewing experiences. Secondly, their reliance on a pinhole camera model, assuming all scene elements are in focus in the input images, presents practical challenges and complicates refocusing during novel-view synthesis. Addressing these limitations, we present a lightweight method based on 3D Gaussian Splatting that utilizes multi-view LDR images of a scene with varying exposure times, apertures, and focus distances as input to reconstruct a high-dynamic-range (HDR) radiance field. By incorporating analytical convolutions of Gaussians based on a thin-lens camera model as well as a tonemapping module, our reconstructions enable the rendering of HDR content with flexible refocusing capabilities. We demonstrate that our combined treatment of HDR and depth of field facilitates real-time cinematic rendering, outperforming the state of the art.

Cellular automata with memory (CAM) are widely used in fields such as image processing, pattern recognition, simulation, and cryptography. The invertibility of CAM is generally considered to be chaotic. Paper [Invertible behavior in elementary cellular automata with memory, Juan C. Seck-Tuoh-Mora et al., Information Sciences, 2012] presented necessary and sufficient conditions for the invertibility of elementary CAM, but it contains a critical error: it classifies identity CAM as non-invertible, whereas identity CAM is undoubtedly invertible. By integrating Amoroso's algorithm and cycle graphs, we provide the correct necessary and sufficient conditions for the invertibility of one-dimensional CAM. Additionally, we link CAM to a specific type of cellular automaton that is isomorphic to CAM, behaves identically, and has easily determinable invertibility. This makes it a promising alternative tool for CAM applications.

Multivariate equivalence testing is needed in a variety of scenarios for drug development. For example, drug products obtained from natural sources may contain many components for which the individual effects and/or their interactions on clinical efficacy and safety cannot be completely characterized. Such lack of sufficient characterization poses a challenge for both generic drug developers to demonstrate and regulatory authorities to determine the sameness of a proposed generic product to its reference product. Another case is to ensure batch-to-batch consistency of naturally derived products containing a vast number of components, such as botanical products. The equivalence or sameness between products containing many components that cannot be individually evaluated needs to be studied in a holistic manner. Multivariate equivalence test based on Mahalanobis distance may be suitable to evaluate many variables holistically. Existing studies based on such method assumed either a predetermined constant margin, for which a consensus is difficult to achieve, or a margin derived from the data, where, however, the randomness is ignored during the testing. In this study, we propose a multivariate equivalence test based on Mahalanobis distance with a data-drive margin with the randomness in the margin considered. Several possible implementations are compared with existing approaches via extensive simulation studies.

Quantum machine learning has demonstrated significant potential in solving practical problems, particularly in statistics-focused areas such as data science and finance. However, challenges remain in preparing and learning statistical models on a quantum processor due to issues with trainability and interpretability. In this letter, we utilize the maximum entropy principle to design a statistics-informed parameterized quantum circuit (SI-PQC) for efficiently preparing and training of quantum computational statistical models, including arbitrary distributions and their weighted mixtures. The SI-PQC features a static structure with trainable parameters, enabling in-depth optimized circuit compilation, exponential reductions in resource and time consumption, and improved trainability and interpretability for learning quantum states and classical model parameters simultaneously. As an efficient subroutine for preparing and learning in various quantum algorithms, the SI-PQC addresses the input bottleneck and facilitates the injection of prior knowledge., Comment: 19 pages, 5 figures

Diffusion models for Text-to-Image (T2I) conditional generation have seen tremendous success recently. Despite their success, accurately capturing user intentions with these models still requires a laborious trial and error process. This challenge is commonly identified as a model alignment problem, an issue that has attracted considerable attention by the research community. Instead of relying on fine-grained linguistic analyses of prompts, human annotation, or auxiliary vision-language models to steer image generation, in this work we present a novel method that relies on an information-theoretic alignment measure. In a nutshell, our method uses self-supervised fine-tuning and relies on point-wise mutual information between prompts and images to define a synthetic training set to induce model alignment. Our comparative analysis shows that our method is on-par or superior to the state-of-the-art, yet requires nothing but a pre-trained denoising network to estimate MI and a lightweight fine-tuning strategy.

Existing efforts are dedicated to designing many topologies and graph-aware strategies for the graph Transformer, which greatly improve the model's representation capabilities. However, manually determining the suitable Transformer architecture for a specific graph dataset or task requires extensive expert knowledge and laborious trials. This paper proposes an evolutionary graph Transformer architecture search framework (EGTAS) to automate the construction of strong graph Transformers. We build a comprehensive graph Transformer search space with the micro-level and macro-level designs. EGTAS evolves graph Transformer topologies at the macro level and graph-aware strategies at the micro level. Furthermore, a surrogate model based on generic architectural coding is proposed to directly predict the performance of graph Transformers, substantially reducing the evaluation cost of evolutionary search. We demonstrate the efficacy of EGTAS across a range of graph-level and node-level tasks, encompassing both small-scale and large-scale graph datasets. Experimental results and ablation studies show that EGTAS can construct high-performance architectures that rival state-of-the-art manual and automated baselines., Comment: This work has been accepted by IEEE Transactions on Neural Networks and Learning Systems. Copyright may be transferred without notice, after which this version may no longer be accessible

Yuan 2.0-M32, with a similar base architecture as Yuan-2.0 2B, uses a mixture-of-experts architecture with 32 experts of which 2 experts are active. A new router network, Attention Router, is proposed and adopted for a more efficient selection of experts, which improves the accuracy compared to the model with classical router network. Yuan 2.0-M32 is trained with 2000B tokens from scratch, and the training computation consumption is only 9.25% of a dense model at the same parameter scale. Yuan 2.0-M32 demonstrates competitive capability on coding, math, and various domains of expertise, with only 3.7B active parameters of 40B in total, and 7.4 GFlops forward computation per token, both of which are only 1/19 of Llama3-70B. Yuan 2.0-M32 surpass Llama3-70B on MATH and ARC-Challenge benchmark, with accuracy of 55.89 and 95.8 respectively. The models and source codes of Yuan 2.0-M32 are released at Github1., Comment: 14 pages,3 figures, 7 tables

The fusion of a low-resolution hyperspectral image (LR-HSI) with a high-resolution multispectral image (HR-MSI) has emerged as an effective technique for achieving HSI super-resolution (SR). Previous studies have mainly concentrated on estimating the posterior distribution of the latent high-resolution hyperspectral image (HR-HSI), leveraging an appropriate image prior and likelihood computed from the discrepancy between the latent HSI and observed images. Low rankness stands out for preserving latent HSI characteristics through matrix factorization among the various priors. However, this method only enhances resolution within the dimensions of the two modalities. To overcome this limitation, we propose a novel continuous low-rank factorization (CLoRF) by integrating two neural representations into the matrix factorization, capturing spatial and spectral information, respectively. This approach enables us to harness both the low rankness from the matrix factorization and the continuity from neural representation in a self-supervised manner. Theoretically, we prove the low-rank property and Lipschitz continuity in the proposed continuous low-rank factorization. Experimentally, our method significantly surpasses existing techniques and achieves user-desired resolutions without the need for neural network retraining.

As the non-fungible token (NFT) market flourishes, price prediction emerges as a pivotal direction for investors gaining valuable insight to maximize returns. However, existing works suffer from a lack of practical definitions and standardized evaluations, limiting their practical application. Moreover, the influence of users' multi-behaviour transactions that are publicly accessible on NFT price is still not explored and exhibits challenges. In this paper, we address these gaps by presenting a practical and hierarchical problem definition. This approach unifies both collection-level and token-level task and evaluation methods, which cater to varied practical requirements of investors. To further understand the impact of user behaviours on the variation of NFT price, we propose a general wallet profiling framework and develop a COmmunity enhanced Multi-bEhavior Transaction graph model, named COMET. COMET profiles wallets with a comprehensive view and considers the impact of diverse relations and interactions within the NFT ecosystem on NFT price variations, thereby improving prediction performance. Extensive experiments conducted in our deployed system demonstrate the superiority of COMET, underscoring its potential in the insight toolkit for NFT investors., Comment: Accepted by KDD 2024 (ADS Track)

In this work, we systematically investigate the efficacy of dynamic activation mechanisms within the LLaMA family of language models. Despite the potential of dynamic activation methods to reduce computation and increase speed in models using the ReLU activation function, our empirical findings have uncovered several inherent pitfalls in the current dynamic activation schemes. Through extensive experiments across various dynamic activation strategies, we demonstrate that LLaMA models usually underperform when compared to their ReLU counterparts, particularly in scenarios demanding high sparsity ratio. We attribute these deficiencies to a combination of factors: 1) the inherent complexity of dynamically predicting activation heads and neurons; 2) the inadequate sparsity resulting from activation functions; 3) the insufficient preservation of information resulting from KV cache skipping. Our analysis not only sheds light on the limitations of dynamic activation in the context of large-scale LLaMA models but also proposes roadmaps for enhancing the design of future sparsity schemes.

As an emerging antenna technology, a fluid antenna system (FAS) enhances spatial diversity to improve both sensing and communication performance by shifting the active antennas among available ports. In this letter, we study the potential of shifting the integrated sensing and communication (ISAC) trade-off with FAS. We propose the model for FAS-enabled ISAC and jointly optimize the transmit beamforming and port selection of FAS. In particular, we aim to minimize the transmit power, while satisfying both communication and sensing requirements. An efficient iterative algorithm based on sparse optimization, convex approximation, and a penalty approach is developed. The simulation results show that the proposed scheme can attain 33% reductions in transmit power with guaranteed sensing and communication performance, showing the great potential of the fluid antenna for striking a flexible tradeoff between sensing and communication in ISAC systems., Comment: 5 pages, 5 figures

When solving systems of banded Toeplitz equations or calculating their inverses, it is necessary to determine the invertibility of the matrices beforehand. In this paper, we equate the invertibility of an $n$-order banded Toeplitz matrix with bandwidth $2k+1$ to that of a small $k*k$ matrix. By utilizing a specially designed algorithm, we compute the invertibility sequence of a class of banded Toeplitz matrices with a time complexity of $5k^2n/2+kn$ and a space complexity of $3k^2$ where $n$ is the size of the largest matrix. This enables efficient preprocessing when solving equation systems and inverses of banded Toeplitz matrices.

This work investigates the effects of complex networks on the collective behavior of a three-state opinion formation model in economic systems. Our model considers two distinct types of investors in financial markets: noise traders and fundamentalists. Financial states evolve via probabilistic dynamics that include economic strategies with local and global influences. The local majoritarian opinion drives noise traders' market behavior, while the market index influences the financial decisions of fundamentalist agents. We introduce a level of market anxiety $q$ present in the decision-making process that influences financial action. In our investigation, nodes of a complex network represent market agents, whereas the links represent their financial interactions. We investigate the stochastic dynamics of the model on three distinct network topologies, including scale-free networks, small-world networks and Erd{\"o}s-R\'enyi random graphs. Our model mirrors various traits observed in real-world financial return series, such as heavy-tailed return distributions, volatility clustering, and short-term memory correlation of returns. The histograms of returns are fitted by coupled Gaussian distributions, quantitatively revealing transitions from a leptokurtic to a mesokurtic regime under specific economic heterogeneity. We show that the market dynamics depend mainly on the average agent connectivity, anxiety level, and market composition rather than on specific features of network topology., Comment: 15 pages, 14 figures

Let $G=(V, E)$ be a graph and let each vertex of $G$ has a lamp and a button. Each button can be of $\sigma^+$-type or $\sigma$-type. Assume that initially some lamps are on and others are off. The button on vertex $x$ is of $\sigma^+$-type ($\sigma$-type, respectively) if pressing the button changes the lamp states on $x$ and on its neighbors in $G$ (the lamp states on the neighbors of $x$ only, respectively). Assume that there is a set $X\subseteq V$ such that pressing buttons on vertices of $X$ lights all lamps on vertices of $G$. In particular, it is known to hold when initially all lamps are off and all buttons are of $\sigma^+$-type. Finding such a set $X$ of the smallest size is NP-hard even if initially all lamps are off and all buttons are of $\sigma^+$-type. Using a linear algebraic approach we design a polynomial-time approximation algorithm for the problem such that for the set $X$ constructed by the algorithm, we have $|X|\le \min\{r,(|V|+{\rm opt})/2\},$ where $r$ is the rank of a (modified) adjacent matrix of $G$ and ${\rm opt}$ is the size of an optimal solution to the problem. To the best of our knowledge, this is the first polynomial-time approximation algorithm for the problem with a nontrivial approximation guarantee.

Large language models (LLMs) have showcased profound capabilities in language understanding and generation, facilitating a wide array of applications. However, there is a notable paucity of detailed, open-sourced methodologies on efficiently scaling LLMs beyond 50 billion parameters with minimum trial-and-error cost and computational resources. In this report, we introduce Tele-FLM (aka FLM-2), a 52B open-sourced multilingual large language model that features a stable, efficient pre-training paradigm and enhanced factual judgment capabilities. Tele-FLM demonstrates superior multilingual language modeling abilities, measured by BPB on textual corpus. Besides, in both English and Chinese foundation model evaluation, it is comparable to strong open-sourced models that involve larger pre-training FLOPs, such as Llama2-70B and DeepSeek-67B. In addition to the model weights, we share the core designs, engineering practices, and training details, which we expect to benefit both the academic and industrial communities.

Hazy images degrade visual quality, and dehazing is a crucial prerequisite for subsequent processing tasks. Most current dehazing methods rely on neural networks and face challenges such as high computational parameter pressure and weak generalization capabilities. This paper introduces PriorNet--a novel, lightweight, and highly applicable dehazing network designed to significantly improve the clarity and visual quality of hazy images while avoiding excessive detail extraction issues. The core of PriorNet is the original Multi-Dimensional Interactive Attention (MIA) mechanism, which effectively captures a wide range of haze characteristics, substantially reducing the computational load and generalization difficulties associated with complex systems. By utilizing a uniform convolutional kernel size and incorporating skip connections, we have streamlined the feature extraction process. Simplifying the number of layers and architecture not only enhances dehazing efficiency but also facilitates easier deployment on edge devices. Extensive testing across multiple datasets has demonstrated PriorNet's exceptional performance in dehazing and clarity restoration, maintaining image detail and color fidelity in single-image dehazing tasks. Notably, with a model size of just 18Kb, PriorNet showcases superior dehazing generalization capabilities compared to other methods. Our research makes a significant contribution to advancing image dehazing technology, providing new perspectives and tools for the field and related domains, particularly emphasizing the importance of improving universality and deployability., Comment: 8 pages, 4 figures

Virtual network embedding (VNE) is an essential resource allocation task in network virtualization, aiming to map virtual network requests (VNRs) onto physical infrastructure. Reinforcement learning (RL) has recently emerged as a promising solution to this problem. However, existing RL-based VNE methods are limited by the unidirectional action design and one-size-fits-all training strategy, resulting in restricted searchability and generalizability. In this paper, we propose a FLexible And Generalizable RL framework for VNE, named FlagVNE. Specifically, we design a bidirectional action-based Markov decision process model that enables the joint selection of virtual and physical nodes, thus improving the exploration flexibility of solution space. To tackle the expansive and dynamic action space, we design a hierarchical decoder to generate adaptive action probability distributions and ensure high training efficiency. Furthermore, to overcome the generalization issue for varying VNR sizes, we propose a meta-RL-based training method with a curriculum scheduling strategy, facilitating specialized policy training for each VNR size. Finally, extensive experimental results show the effectiveness of FlagVNE across multiple key metrics. Our code is available at GitHub (https://github.com/GeminiLight/flag-vne)., Comment: Accepted by IJCAI 2024

One-on-one help from a teacher is highly impactful for students, yet extremely challenging to support in massive online courses (MOOCs). In this work, we present TeachNow: a novel system that lets volunteer teachers from anywhere in the world instantly provide 1:1 help sessions to students in MOOCs, without any scheduling or coordination overhead. TeachNow works by quickly finding an online student to help and putting them in a collaborative working session with the teacher. The spontaneous, on-demand nature of TeachNow gives teachers the flexibility to help whenever their schedule allows. We share our experiences deploying TeachNow as an experimental feature in a six week online CS1 course with 9,000 students and 600 volunteer teachers. Even as an optional activity, TeachNow was used by teachers to provide over 12,300 minutes of 1:1 help to 375 unique students. Through a carefully designed randomised control trial, we show that TeachNow sessions increased student course retention rate by almost 15%. Moreover, the flexibility of our system captured valuable volunteer time that would otherwise go to waste. Lastly, TeachNow was rated by teachers as one of the most enjoyable and impactful aspects of their involvement in the course. We believe TeachNow is an important step towards providing more human-centered support in massive online courses.

Banded Toeplitz matrices over $\mathbb{F}_p$, as a well-known class of matrices, have been extensively studied in the fields of coding theory and automata theory. In this paper, we discover that both determinants and inverses of banded Toeplitz matrices over $\mathbb{F}_p$ exhibit periodicity. For a Toeplitz matrix with bandwidth $k$, The period $P(f)$ is related to the parameters on the band and is independent of the order, with an upper limit of $P(f) \le p^{k-1}-1$. We provide an algorithm which can compute the determinant of any order banded Toeplitz matrix within $O(k^4)$. And its inverse can be represented by three submatrices of size $P(f)*P(f)$ located respectively on the diagonal, above the diagonal, and below the diagonal. Thus, the computational cost for calculating the inverse is fixed, and our algorithm can solve it within $O(k^5)+3kP(f)^2$. This is the first time that the periodicity of determinants and inverses of general banded Toeplitz matrices over $\mathbb{F}_p$ has been computed and proven.

Using the tight-binding model, a multi-terminal superconductor(S) device is proposed, where the structures of the center region are primitive grpahene(G) and Y-shaped Kekul\'{e} graphene superlattice(GS), respectively. The intravalley Andreev reflection is studied in this model through the utilization of the non-equilibrium Green's function method. In the G/S device, due to the time-reverse symmetry the dominant process of Andreev reflection is intervalley reflection. In a three-terminal GS/S device, it has been observed that the coefficient of intravalley Andreev reflection can surpass that of intervalley reflection in crossed Andreev reflection. This is attributed to the coupling between valleys K and -K within the first Brillouin zone, resulting in enhanced intervalley scattering. The valley-dependent transport of Andreev reflection can be influenced by the phase difference between superconductor terminals. The valley polarization of the local Andreev reflection and the crossed Andreev reflection could be controlled by adjusting the structure of the central region.

Detecting transmission towers from synthetic aperture radar (SAR) images remains a challenging task due to the comparatively small size and side-looking geometry, with background clutter interference frequently hindering tower identification. A large number of interfering signals superimposes the return signal from the tower. We found that localizing or prompting positions of power transmission towers is beneficial to address this obstacle. Based on this revelation, this paper introduces prompt learning into the oriented object detector (P2Det) for multimodal information learning. P2Det contains the sparse prompt coding and cross-attention between the multimodal data. Specifically, the sparse prompt encoder (SPE) is proposed to represent point locations, converting prompts into sparse embeddings. The image embeddings are generated through the Transformer layers. Then a two-way fusion module (TWFM) is proposed to calculate the cross-attention of the two different embeddings. The interaction of image-level and prompt-level features is utilized to address the clutter interference. A shape-adaptive refinement module (SARM) is proposed to reduce the effect of aspect ratio. Extensive experiments demonstrated the effectiveness of the proposed model on high-resolution SAR images. P2Det provides a novel insight for multimodal object detection due to its competitive performance., Comment: 22 pages, 12figures