Search

The rapid evolution of mobile edge computing (MEC) has introduced significant challenges in optimizing resource allocation in highly dynamic wireless communication systems, in which task offloading decisions should be made in real-time. However, existing resource allocation strategies cannot well adapt to the dynamic and heterogeneous characteristics of MEC systems, since they are short of scalability, context-awareness, and interpretability. To address these issues, this paper proposes a novel retrieval-augmented generation (RAG) method to improve the performance of MEC systems. Specifically, a latency minimization problem is first proposed to jointly optimize the data offloading ratio, transmit power allocation, and computing resource allocation. Then, an LLM-enabled information-retrieval mechanism is proposed to solve the problem efficiently. Extensive experiments across multi-user, multi-task, and highly dynamic offloading scenarios show that the proposed method consistently reduces latency compared to several DL-based approaches, achieving 57% improvement under varying user computing ability, 86% with different servers, 30% under distinct transmit powers, and 42% for varying data volumes. These results show the effectiveness of LLM-driven solutions to solve the resource allocation problems in MEC systems., Comment: This manuscript has been submitted to IEEE

As an emerging technology, cooperative bi-static integrated sensing and communication (ISAC) is promising to achieve high-precision sensing, high-rate communication as well as self-interference (SI) avoidance. This paper investigates the two-timescale design for access point (AP) mode selection to realize the full potential of the cooperative bi-static ISAC network with low system overhead, where the beamforming at the APs is adapted to the rapidly-changing instantaneous channel state information (CSI), while the AP mode is adapted to the slowly-changing statistical CSI. We first apply the minimum mean square error (MMSE) estimator to estimate the channel between the APs and the channels from the APs to the user equipments (UEs). Then we adopt the low-complexity maximum ratio transmission (MRT) beamforming and the maximum ratio combining (MRC) detector, and derive the closed-form expressions of the communication rate and the sensing signal-to-interference-plus-noise-ratio (SINR). We formulate a non-convex integer optimization problem to maximize the minimum sensing SINR under the communication quality of service (QoS) constraints. McCormick envelope relaxation and successive convex approximation (SCA) techniques are applied to solve the challenging non-convex integer optimization problem. Simulation results validate the closed-form expressions and prove the convergence and effectiveness of the proposed AP mode selection scheme., Comment: 13 pages, 8 figures

We present DeepSeek-V3, a strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token. To achieve efficient inference and cost-effective training, DeepSeek-V3 adopts Multi-head Latent Attention (MLA) and DeepSeekMoE architectures, which were thoroughly validated in DeepSeek-V2. Furthermore, DeepSeek-V3 pioneers an auxiliary-loss-free strategy for load balancing and sets a multi-token prediction training objective for stronger performance. We pre-train DeepSeek-V3 on 14.8 trillion diverse and high-quality tokens, followed by Supervised Fine-Tuning and Reinforcement Learning stages to fully harness its capabilities. Comprehensive evaluations reveal that DeepSeek-V3 outperforms other open-source models and achieves performance comparable to leading closed-source models. Despite its excellent performance, DeepSeek-V3 requires only 2.788M H800 GPU hours for its full training. In addition, its training process is remarkably stable. Throughout the entire training process, we did not experience any irrecoverable loss spikes or perform any rollbacks. The model checkpoints are available at https://github.com/deepseek-ai/DeepSeek-V3.

In this report, we introduce Qwen2.5, a comprehensive series of large language models (LLMs) designed to meet diverse needs. Compared to previous iterations, Qwen 2.5 has been significantly improved during both the pre-training and post-training stages. In terms of pre-training, we have scaled the high-quality pre-training datasets from the previous 7 trillion tokens to 18 trillion tokens. This provides a strong foundation for common sense, expert knowledge, and reasoning capabilities. In terms of post-training, we implement intricate supervised finetuning with over 1 million samples, as well as multistage reinforcement learning. Post-training techniques enhance human preference, and notably improve long text generation, structural data analysis, and instruction following. To handle diverse and varied use cases effectively, we present Qwen2.5 LLM series in rich sizes. Open-weight offerings include base and instruction-tuned models, with quantized versions available. In addition, for hosted solutions, the proprietary models currently include two mixture-of-experts (MoE) variants: Qwen2.5-Turbo and Qwen2.5-Plus, both available from Alibaba Cloud Model Studio. Qwen2.5 has demonstrated top-tier performance on a wide range of benchmarks evaluating language understanding, reasoning, mathematics, coding, human preference alignment, etc. Specifically, the open-weight flagship Qwen2.5-72B-Instruct outperforms a number of open and proprietary models and demonstrates competitive performance to the state-of-the-art open-weight model, Llama-3-405B-Instruct, which is around 5 times larger. Qwen2.5-Turbo and Qwen2.5-Plus offer superior cost-effectiveness while performing competitively against GPT-4o-mini and GPT-4o respectively. Additionally, as the foundation, Qwen2.5 models have been instrumental in training specialized models such as Qwen2.5-Math, Qwen2.5-Coder, QwQ, and multimodal models.

Fuel efficiency is a crucial aspect of long-distance cargo transportation by oil-powered trucks that economize on costs and decrease carbon emissions. Current predictive control methods depend on an accurate model of vehicle dynamics and engine, including weight, drag coefficient, and the Brake-specific Fuel Consumption (BSFC) map of the engine. We propose a pure data-driven method, Neural Predictive Control (NPC), which does not use any physical model for the vehicle. After training with over 20,000 km of historical data, the novel proposed NVFormer implicitly models the relationship between vehicle dynamics, road slope, fuel consumption, and control commands using the attention mechanism. Based on the online sampled primitives from the past of the current freight trip and anchor-based future data synthesis, the NVFormer can infer optimal control command for reasonable fuel consumption. The physical model-free NPC outperforms the base PCC method with 2.41% and 3.45% more significant fuel saving in simulation and open-road highway testing, respectively., Comment: 7 pages, 6 figures, for associated mpeg file, see https://www.youtube.com/watch?v=hqgpj7LhiL4

As large language models (LLMs) grow in size, traditional full fine-tuning becomes increasingly impractical due to its high computational and storage costs. Although popular parameter-efficient fine-tuning methods, such as LoRA, have significantly reduced the number of tunable parameters, there is still room for further optimization. In this work, we propose ASLoRA, a cross-layer parameter-sharing strategy combining global sharing with partial adaptive sharing. Specifically, we share the low-rank matrix A across all layers and adaptively merge matrix B during training. This sharing mechanism not only mitigates overfitting effectively but also captures inter-layer dependencies, significantly enhancing the model's representational capability. We conduct extensive experiments on various NLP tasks, showing that ASLoRA outperforms LoRA while using less than 25% of the parameters, highlighting its flexibility and superior parameter efficiency. Furthermore, in-depth analyses of the adaptive sharing strategy confirm its significant advantages in enhancing both model flexibility and task adaptability.

Human emotion synthesis is a crucial aspect of affective computing. It involves using computational methods to mimic and convey human emotions through various modalities, with the goal of enabling more natural and effective human-computer interactions. Recent advancements in generative models, such as Autoencoders, Generative Adversarial Networks, Diffusion Models, Large Language Models, and Sequence-to-Sequence Models, have significantly contributed to the development of this field. However, there is a notable lack of comprehensive reviews in this field. To address this problem, this paper aims to address this gap by providing a thorough and systematic overview of recent advancements in human emotion synthesis based on generative models. Specifically, this review will first present the review methodology, the emotion models involved, the mathematical principles of generative models, and the datasets used. Then, the review covers the application of different generative models to emotion synthesis based on a variety of modalities, including facial images, speech, and text. It also examines mainstream evaluation metrics. Additionally, the review presents some major findings and suggests future research directions, providing a comprehensive understanding of the role of generative technology in the nuanced domain of emotion synthesis., Comment: 25 pages, 10 figures

This study proposes the IoT-Enhanced Pose Optimization Network (IE-PONet) for high-precision 3D pose estimation and motion optimization of track and field athletes. IE-PONet integrates C3D for spatiotemporal feature extraction, OpenPose for real-time keypoint detection, and Bayesian optimization for hyperparameter tuning. Experimental results on NTURGB+D and FineGYM datasets demonstrate superior performance, with AP\(^p50\) scores of 90.5 and 91.0, and mAP scores of 74.3 and 74.0, respectively. Ablation studies confirm the essential roles of each module in enhancing model accuracy. IE-PONet provides a robust tool for athletic performance analysis and optimization, offering precise technical insights for training and injury prevention. Future work will focus on further model optimization, multimodal data integration, and developing real-time feedback mechanisms to enhance practical applications., Comment: 17 pages

Cancer, with its inherent heterogeneity, is commonly categorized into distinct subtypes based on unique traits, cellular origins, and molecular markers specific to each type. However, current studies primarily rely on complete multi-omics datasets for predicting cancer subtypes, often overlooking predictive performance in cases where some omics data may be missing and neglecting implicit relationships across multiple layers of omics data integration. This paper introduces Multi-Layer Matrix Factorization (MLMF), a novel approach for cancer subtyping that employs multi-omics data clustering. MLMF initially processes multi-omics feature matrices by performing multi-layer linear or nonlinear factorization, decomposing the original data into latent feature representations unique to each omics type. These latent representations are subsequently fused into a consensus form, on which spectral clustering is performed to determine subtypes. Additionally, MLMF incorporates a class indicator matrix to handle missing omics data, creating a unified framework that can manage both complete and incomplete multi-omics data. Extensive experiments conducted on 10 multi-omics cancer datasets, both complete and with missing values, demonstrate that MLMF achieves results that are comparable to or surpass the performance of several state-of-the-art approaches.

Youth Participatory Action Research (YPAR) is traditionally valued for its developmental benefits for youth, such as enhancing leadership and critical thinking skills. However, this brief evidences that YPAR's potential extends significantly beyond individual growth, serving as a powerful mechanism for organizational development and academic field-building. We explore how YPAR facilitates local insights into organizational practices from a youth perspective and contributes uniquely to the broader academic discourse, effectively bridging the research-practice gap.

A search for $\eta_c(2S)\to p\bar{p}K^+K^-$, together with measurement of branching fractions of $\chi_{cJ(J=0,1,2)}\to p\bar{p}K^+K^-$ in the $\psi(3686) \to \gamma \eta_c(2S)$ and the $\psi(3686) \to \gamma \chi_{cJ}$ radiative decays, is performed with $(2712.4\pm14.3)\times 10^6$ $\psi(3686)$ events collected with the BESIII detector at the BEPCII collider. An evidence for $\eta_c(2S)\to p\bar{p}K^+K^-$ is found, with a significance of $3.3\sigma$. The product branching fraction of $\mathcal{B}[\psi(3686)\to\gamma\eta_c(2S)]\cdot\mathcal{B}[\eta_c(2S)\to p\bar{p}K^+K^-]$ is determined to be $(1.98\mkern 2mu\pm\mkern 2mu0.41_{\text{stat.}}\mkern 2mu\pm\mkern 2mu0.99_{\text{syst.}})\times 10^{-7}$. The product branching fractions of $\mathcal{B}[\psi(3686)\to\gamma\chi_{cJ}]\cdot\mathcal{B}[\chi_{cJ}\to p\bar{p}K^+K^-]$ are measured to be $(2.49\mkern 2mu\pm\mkern 2mu 0.03_{\text{stat.}}\mkern 2mu\pm\mkern 2mu 0.15_{\text{syst.}})\times 10^{-5}$, $(1.83\mkern 2mu \pm\mkern 2mu 0.02_{\text{stat.}}\mkern 2mu \pm\mkern 2mu 0.11_{\text{syst.}})\times 10^{-5}$, and $(2.43\mkern 2mu\pm\mkern 2mu 0.02_{\text{stat.}}\mkern 2mu\pm\mkern 2mu 0.15_{\text{syst.}})\times 10^{-5}$, for $J=0,\ 1$, and 2, respectively., Comment: 12 pages, 2 figures

Cross-modal hashing (CMH) has appeared as a popular technique for cross-modal retrieval due to its low storage cost and high computational efficiency in large-scale data. Most existing methods implicitly assume that multi-modal data is correctly labeled, which is expensive and even unattainable due to the inevitable imperfect annotations (i.e., noisy labels) in real-world scenarios. Inspired by human cognitive learning, a few methods introduce self-paced learning (SPL) to gradually train the model from easy to hard samples, which is often used to mitigate the effects of feature noise or outliers. It is a less-touched problem that how to utilize SPL to alleviate the misleading of noisy labels on the hash model. To tackle this problem, we propose a new cognitive cross-modal retrieval method called Robust Self-paced Hashing with Noisy Labels (RSHNL), which can mimic the human cognitive process to identify the noise while embracing robustness against noisy labels. Specifically, we first propose a contrastive hashing learning (CHL) scheme to improve multi-modal consistency, thereby reducing the inherent semantic gap. Afterward, we propose center aggregation learning (CAL) to mitigate the intra-class variations. Finally, we propose Noise-tolerance Self-paced Hashing (NSH) that dynamically estimates the learning difficulty for each instance and distinguishes noisy labels through the difficulty level. For all estimated clean pairs, we further adopt a self-paced regularizer to gradually learn hash codes from easy to hard. Extensive experiments demonstrate that the proposed RSHNL performs remarkably well over the state-of-the-art CMH methods., Comment: 9 pages, AAAI 25 conference

We introduce EnerVerse, a comprehensive framework for embodied future space generation specifically designed for robotic manipulation tasks. EnerVerse seamlessly integrates convolutional and bidirectional attention mechanisms for inner-chunk space modeling, ensuring low-level consistency and continuity. Recognizing the inherent redundancy in video data, we propose a sparse memory context combined with a chunkwise unidirectional generative paradigm to enable the generation of infinitely long sequences. To further augment robotic capabilities, we introduce the Free Anchor View (FAV) space, which provides flexible perspectives to enhance observation and analysis. The FAV space mitigates motion modeling ambiguity, removes physical constraints in confined environments, and significantly improves the robot's generalization and adaptability across various tasks and settings. To address the prohibitive costs and labor intensity of acquiring multi-camera observations, we present a data engine pipeline that integrates a generative model with 4D Gaussian Splatting (4DGS). This pipeline leverages the generative model's robust generalization capabilities and the spatial constraints provided by 4DGS, enabling an iterative enhancement of data quality and diversity, thus creating a data flywheel effect that effectively narrows the sim-to-real gap. Finally, our experiments demonstrate that the embodied future space generation prior substantially enhances policy predictive capabilities, resulting in improved overall performance, particularly in long-range robotic manipulation tasks., Comment: Website: https://sites.google.com/view/enerverse

Debris disks, which consist of dust, planetesimals, planets, and gas, offer a unique window into the mineralogical composition of their parent bodies, especially during the critical phase of terrestrial planet formation spanning 10 to a few hundred million years. Observations from the $\textit{Spitzer}$ Space Telescope have unveiled thousands of debris disks, yet systematic studies remain scarce, let alone those with unsupervised clustering techniques. This study introduces $\texttt{CLUES}$ (CLustering UnsupErvised with Sequencer), a novel, non-parametric, fully-interpretable machine-learning spectral analysis tool designed to analyze and classify the spectral data of debris disks. $\texttt{CLUES}$ combines multiple unsupervised clustering methods with multi-scale distance measures to discern new groupings and trends, offering insights into compositional diversity and geophysical processes within these disks. Our analysis allows us to explore a vast parameter space in debris disk mineralogy and also offers broader applications in fields such as protoplanetary disks and solar system objects. This paper details the methodology, implementation, and initial results of $\texttt{CLUES}$, setting the stage for more detailed follow-up studies focusing on debris disk mineralogy and demographics., Comment: 23 pages, 16 figures, Accepted to ApJS, $\texttt{CLUES}$ software available on GitHub

With the increasing code reasoning capabilities of existing large language models (LLMs) and breakthroughs in reasoning models like OpenAI o1 and o3, there is a growing need to develop more challenging and comprehensive benchmarks that effectively test their sophisticated competition-level coding abilities. Existing benchmarks, like LiveCodeBench and USACO, fall short due to the unavailability of private test cases, lack of support for special judges, and misaligned execution environments. To bridge this gap, we introduce CodeElo, a standardized competition-level code generation benchmark that effectively addresses all these challenges for the first time. CodeElo benchmark is mainly based on the official CodeForces platform and tries to align with the platform as much as possible. We compile the recent six months of contest problems on CodeForces with detailed information such as contest divisions, problem difficulty ratings, and problem algorithm tags. We introduce a unique judging method in which problems are submitted directly to the platform and develop a reliable Elo rating calculation system that aligns with the platform and is comparable with human participants but has lower variance. By testing on our CodeElo, we provide the Elo ratings of 30 existing popular open-source and 3 proprietary LLMs for the first time. The results show that o1-mini and QwQ-32B-Preview stand out significantly, achieving Elo ratings of 1578 and 1261, respectively, while other models struggle even with the easiest problems, placing in the lowest 25 percent among all human participants. Detailed analysis experiments are also conducted to provide insights into performance across algorithms and comparisons between using C++ and Python, which can suggest directions for future studies.

Privacy-preserving machine learning (PPML) enables clients to collaboratively train deep learning models without sharing private datasets, but faces privacy leakage risks due to gradient leakage attacks. Prevailing methods leverage secure aggregation strategies to enhance PPML, where clients leverage masks and secret sharing to further protect gradient data while tolerating participant dropouts. These methods, however, require frequent inter-client communication to negotiate keys and perform secret sharing, leading to substantial communication overhead. To tackle this issue, we propose NET-SA, an efficient secure aggregation architecture for PPML based on in-network computing. NET-SA employs seed homomorphic pseudorandom generators for local gradient masking and utilizes programmable switches for seed aggregation. Accurate and secure gradient aggregation is then performed on the central server based on masked gradients and aggregated seeds. This design effectively reduces communication overhead due to eliminating the communication-intensive phases of seed agreement and secret sharing, with enhanced dropout tolerance due to overcoming the threshold limit of secret sharing. Extensive experiments on server clusters and Intel Tofino programmable switch demonstrate that NET-SA achieves up to 77x and 12x enhancements in runtime and 2x decrease in total client communication cost compared with state-of-the-art methods., Comment: 10 pages, 8 figures, 3 tables

AI agents have become increasingly prevalent in recent years, driven by significant advancements in the field of large language models (LLMs). Mobile GUI agents, a subset of AI agents, are designed to autonomously perform tasks on mobile devices. While numerous studies have introduced agents, datasets, and benchmarks to advance mobile GUI agent research, many existing datasets focus on static frame evaluations and fail to provide a comprehensive platform for assessing performance on real-world, in-the-wild tasks. To address this gap, we present Android Agent Arena (A3), a novel evaluation platform. Unlike existing in-the-wild systems, A3 offers: (1) meaningful and practical tasks, such as real-time online information retrieval and operational instructions; (2) a larger, more flexible action space, enabling compatibility with agents trained on any dataset; and (3) automated business-level LLM-based evaluation process. A3 includes 21 widely used general third-party apps and 201 tasks representative of common user scenarios, providing a robust foundation for evaluating mobile GUI agents in real-world situations and a new autonomous evaluation process for less human labor and coding expertise. The project is available at \url{https://yuxiangchai.github.io/Android-Agent-Arena/}.

Semi-supervised domain adaptation (SSDA) has been extensively researched due to its ability to improve classification performance and generalization ability of models by using a small amount of labeled data on the target domain. However, existing methods cannot effectively adapt to the target domain due to difficulty in fully learning rich and complex target semantic information and relationships. In this paper, we propose a novel SSDA learning framework called semantic regularization learning (SERL), which captures the target semantic information from multiple perspectives of regularization learning to achieve adaptive fine-tuning of the source pre-trained model on the target domain. SERL includes three robust semantic regularization techniques. Firstly, semantic probability contrastive regularization (SPCR) helps the model learn more discriminative feature representations from a probabilistic perspective, using semantic information on the target domain to understand the similarities and differences between samples. Additionally, adaptive weights in SPCR can help the model learn the semantic distribution correctly through the probabilities of different samples. To further comprehensively understand the target semantic distribution, we introduce hard-sample mixup regularization (HMR), which uses easy samples as guidance to mine the latent target knowledge contained in hard samples, thereby learning more complete and complex target semantic knowledge. Finally, target prediction regularization (TPR) regularizes the target predictions of the model by maximizing the correlation between the current prediction and the past learned objective, thereby mitigating the misleading of semantic information caused by erroneous pseudo-labels. Extensive experiments on three benchmark datasets demonstrate that our SERL method achieves state-of-the-art performance.

This Letter reports the precise measurement of reactor antineutrino spectrum and flux based on the full data set of 4.7 million inverse-beta-decay (IBD) candidates collected at Daya Bay near detectors. Expressed in terms of the IBD yield per fission, the antineutrino spectra from all reactor fissile isotopes and the specific $\mathrm{^{235}U}$ and $\mathrm{^{239}Pu}$ isotopes are measured with 1.3$\%$, 3$\%$ and 8$\%$ uncertainties respectively near the 3 MeV spectrum peak in reconstructed energy, reaching the best precision in the world. The total antineutrino flux and isotopic $\mathrm{^{235}U}$ and $\mathrm{^{239}Pu}$ fluxes are precisely measured to be $5.84\pm0.07$, $6.16\pm0.12$ and $4.16\pm0.21$ in units of $10^{-43} \mathrm{cm^2/fission}$. These measurements are compared with the Huber-Mueller (HM) model, the reevaluated conversion model based on the Kurchatov Institute (KI) measurement and the latest Summation Model (SM2023). The Daya Bay flux shows good consistency with KI and SM2023 models, but disagrees with HM model. The Daya Bay spectrum, however, disagrees with all model predictions.

The rotational Doppler effect (RDE) of structured light carrying orbital angular momentum (OAM) has attracted widespread attention for applications in optical sensors and OAM spectrum detection. These studies, however, based on RDE, are mostly focused on the motion parameters of rotating objects; other equally important attitude characteristics, e.g., the tilt angle of the axis of rotation, have rarely been considered. We observed an interesting phenomenon in the experiments: the rotational Doppler spectral distribution varies with the ellipticity of the elliptical optical vortex (EOV) and the tilt angle between the rotational axis and optical axis, which inspired us to wonder if it is possible to detect oblique rotational axis or compensate the rotational Doppler broadening effect induced by oblique incidence by utilizing the EOV. Here, we reveal the RDE quantitative relationship with tilt angle and ellipticity for the first time and report a novel approach for tilt angle measurement. By employing a series of EOV with periodically varying ellipticity to illuminate a rotating object and analyze the time-frequency spectral distribution of scattered light associated with ellipticity and tilt angle, the tilt angle can be acquired accurately based on the specific relationship between the tilt angle and ellipticity of the EOV. Furthermore, the spectrum broadening effect arising from oblique incidence in the actual scenario may be addressed through our scheme. The method may find applications in industrial manufacturing and target attitude measurement, and our results provide new insights for obtaining more information about objects.

Large Language Models (LLMs) have demonstrated remarkable performance across a wide range of tasks by understanding input information and predicting corresponding outputs. However, the internal mechanisms by which LLMs comprehend input and make effective predictions remain poorly understood. In this paper, we explore the working mechanism of LLMs in information processing from the perspective of Information Bottleneck Theory. We propose a non-training construction strategy to define a task space and identify the following key findings: (1) LLMs compress input information into specific task spaces (e.g., sentiment space, topic space) to facilitate task understanding; (2) they then extract and utilize relevant information from the task space at critical moments to generate accurate predictions. Based on these insights, we introduce two novel approaches: an Information Compression-based Context Learning (IC-ICL) and a Task-Space-guided Fine-Tuning (TS-FT). IC-ICL enhances reasoning performance and inference efficiency by compressing retrieved example information into the task space. TS-FT employs a space-guided loss to fine-tune LLMs, encouraging the learning of more effective compression and selection mechanisms. Experiments across multiple datasets validate the effectiveness of task space construction. Additionally, IC-ICL not only improves performance but also accelerates inference speed by over 40\%, while TS-FT achieves superior results with a minimal strategy adjustment., Comment: 9 pages, 9 figures, 3 tables

This paper considers reconfigurable intelligent surface (RIS)-aided integrated sensing and communication (ISAC) systems under dual-polarized (DP) channels. Unlike the existing ISAC systems, which ignored polarization of electromagnetic waves, this study adopts DP base station (BS) and DP RIS to serve users with a pair of DP antennas. The achievable sum rate is maximized through jointly optimizing the beamforming matrix at the DP BS, and the reflecting coefficients at the DP RIS. To address this problem, we first utilize the weighted minimum mean-square error (WMMSE) method to transform the objective function into a more tractable form, and then an alternating optimization (AO) method is employed to decouple the original problem into two subproblems. Due to the constant modulus constraint, the DP RIS reflection matrix optimization problem is addressed by the majorization-minimization (MM) method. For the DP beamforming matrix, we propose a penalty-based algorithm that can obtain a low-complexity closed-form solution. Simulation results validate the advantage of deploying DP transmit array and DP RIS in the considered ISAC systems.

To date, nearly all the reported Lamb wave resonators (LWRs) are straight crested LWRs, which suffer from inherent spurious modes and low quality factors (Q). For the first time, this work demonstrates a circular crested LWR. Its advantages over the straight crested LWR are presented comprehensively by studying their fundamental symmetric (S0) mode, which is the simplest and most representative Lamb wave mode. Utilizing circular crested Lamb waves, the proposed resonator avoids only utilizing waves propagating in the lateral direction in the straight crested LWRs, thus eliminating the transverse spurious modes as no transverse direction exists. Besides, different from straight crested Lamb waves maintaining the same displacement amplitude along the propagation direction, circular crested Lamb waves exhibit displacement attenuation toward device edges, which effectively concentrates energy in the device center, assists in reducing energy loss through anchors, and improves Q. Based on 20 at.% scandium-doped aluminum nitride (Al0.8Sc0.2N) thin films, the microfabricated circular crested LWR effectively suppresses transverse spurious modes and achieves a 40.7% Q improvement in experiments with no degradation in effective electromechanical coupling coefficients (keff2) compared with the conventional straight crested LWR when working in S0 mode in experiments, in contrast with the conventional straight crested LWR which shows inherent experimental transverse spurious modes. Moreover, the free edges covered by top electrodes enhance device robustness against misalignment and over-etching in the fabrication process. With the advantages of spurious mode suppression, Q enhancement, and fabrication robustness, the circular crested LWR is a promising candidate for next-generation filters and oscillators., Comment: 23 pages, 13 figures

Data-centric methods have shown great potential in understanding and predicting spatiotemporal dynamics, enabling better design and control of the object system. However, deep learning models often lack interpretability, fail to obey intrinsic physics, and struggle to cope with the various domains. While geometry-based methods, e.g., graph neural networks (GNNs), have been proposed to further tackle these challenges, they still need to find the implicit physical laws from large datasets and rely excessively on rich labeled data. In this paper, we herein introduce the conservation-informed GNN (CiGNN), an end-to-end explainable learning framework, to learn spatiotemporal dynamics based on limited training data. The network is designed to conform to the general conservation law via symmetry, where conservative and non-conservative information passes over a multiscale space enhanced by a latent temporal marching strategy. The efficacy of our model has been verified in various spatiotemporal systems based on synthetic and real-world datasets, showing superiority over baseline models. Results demonstrate that CiGNN exhibits remarkable accuracy and generalizability, and is readily applicable to learning for prediction of various spatiotemporal dynamics in a spatial domain with complex geometry.

Hard Attention Mechanisms (HAMs) effectively filter essential information discretely and significantly boost the performance of machine learning models on large datasets. Nevertheless, they confront the challenge of non-differentiability, which raises the risk of convergence to a local optimum. Quantum Annealing (QA) is expected to solve the above dilemma. We propose a Quantum Annealing Hard Attention Mechanism (QAHAM) for faster convergence to the global optimum without the need to compute gradients by exploiting the quantum tunneling effect. Based on the above theory, we construct a Quantum Annealing Hard Attention Network (QAHAN) on D-Wave and Pytorch platforms for MNIST and CIFAR-10 multi-classification. Experimental results indicate that the QAHAN converges faster, exhibits smoother accuracy and loss curves, and demonstrates superior noise robustness compared to two traditional HAMs. Predictably, our scheme accelerates the convergence between the fields of quantum algorithms and machine learning, while advancing the field of quantum machine vision.

Event cameras are biologically inspired sensors that emit events asynchronously with remarkable temporal resolution, garnering significant attention from both industry and academia. Mainstream methods favor frame and voxel representations, which reach a satisfactory performance while introducing time-consuming transformation, bulky models, and sacrificing fine-grained temporal information. Alternatively, Point Cloud representation demonstrates promise in addressing the mentioned weaknesses, but it ignores the polarity information, and its models have limited proficiency in abstracting long-term events' features. In this paper, we propose a frequency-aware network named FECNet that leverages Event Cloud representations. FECNet fully utilizes 2S-1T-1P Event Cloud by innovating the event-based Group and Sampling module. To accommodate the long sequence events from Event Cloud, FECNet embraces feature extraction in the frequency domain via the Fourier transform. This approach substantially extinguishes the explosion of Multiply Accumulate Operations (MACs) while effectively abstracting spatial-temporal features. We conducted extensive experiments on event-based object classification, action recognition, and human pose estimation tasks, and the results substantiate the effectiveness and efficiency of FECNet., Comment: Under Review

In open-ended generative tasks like narrative writing or dialogue, large language models often exhibit cultural biases, showing limited knowledge and generating templated outputs for less prevalent cultures. Recent works show that these biases may stem from uneven cultural representation in pretraining corpora. This work investigates how pretraining leads to biased culture-conditioned generations by analyzing how models associate entities with cultures based on pretraining data patterns. We propose the MEMOed framework (MEMOrization from pretraining document) to determine whether a generation for a culture arises from memorization. Using MEMOed on culture-conditioned generations about food and clothing for 110 cultures, we find that high-frequency cultures in pretraining data yield more generations with memorized symbols, while some low-frequency cultures produce none. Additionally, the model favors generating entities with extraordinarily high frequency regardless of the conditioned culture, reflecting biases toward frequent pretraining terms irrespective of relevance. We hope that the MEMOed framework and our insights will inspire more works on attributing model performance on pretraining data.

Accurate description of deformed atomic nuclei by the orbital-free density functional theory has been a longstanding textbook challenge, due to the difficulty in accounting for the intricate quantum shell effects that are present in such systems. Orbital-free density functional theory is, in principle, capable of describing all effects of nuclear systems, as guaranteed by the Hohenberg-Kohn theorem. However, from a microscopic perspective, shell and deformation effects are believed to be intrinsically connected to single-orbital structures, posing a significant challenge for orbital-free approaches. Here, we develop a machine learning approach to the orbital-free density functional theory, which is capable of achieving a high level of accuracy in describing the ground-state properties and potential energy curves for both spherical $^{16}$O and deformed $^{20}$Ne nuclei. This is the inaugural instance where a fully orbital-free energy density functional has succeeded in taming the complex shell effects in deformed nuclei. It demonstrates that the orbital-free energy density functional, which is directly based on the Hohenberg-Kohn theorem, is not only a theoretical concept but also a practical one for nuclear systems., Comment: 13 pages, 3 figures

While ridges in the scalogram, determined by the squared modulus of analytic wavelet transform (AWT), have been widely applied in time series analysis and time-frequency analysis, their behavior in noisy environments remains underexplored. We fill in this gap in this paper. We define ridges as paths formed by local maximizers of the scalogram along the scale axis, and analyze their properties when the signal is contaminated by stationary Gaussian noise. In addition to establishing several key properties of the AWT for random processes, we investigate the probabilistic characteristics of the resulting random ridge points in the scalogram. Specifically, we establish the uniqueness property of the ridge point at individual time instances and prove the upper hemicontinuity of the set of ridge points. Furthermore, we derive bounds on the probability that the deviation between the ridge points of noisy and clean signals exceeds a specified threshold. These bounds are expressed in terms of constants related to the signal-to-noise ratio and apply when the signal satisfies the adaptive harmonic model and is corrupted by a stationary Gaussian process. To achieve these results, we derive maximal inequalities for the complex modulus of nonstationary Gaussian processes, leveraging classical tools such as the Borell-TIS inequality and Dudley's theorem, which might be of independent interest.

Vertically parametrised polynomial systems are a particular nice class of parametrised polynomial systems for which a lot of interesting algebraic information is encoded in its combinatorics. Given a fixed polynomial system, we empirically study what constitutes a good vertically parametrised polynomial system that gives rise to it and how to construct said vertically parametrised polynomial system. For data, we use all polynomial systems in ODEbase, which we have transcribed to an OSCAR readable format, and made available as a Julia package OscarODEbase., Comment: 14 pages, 3 figures

The changing-look blazars (CLBs) are the blazars that their optical spectral lines at different epochs show a significant changes and present a clear transition between the standard FSRQ and BL Lac types. The changing-look phenomena in blazars are highly significant for enhancing our understanding of certain physical problems of active galactic nuclei (AGNs), such as the potential mechanism of the state transition in the accretion process of the supermassive black holes in the central engine of AGNs, the possible intrinsic variation of the jet, and the connection between the accretion disk and the jet. Currently, the CLBs reported in the literature are still rare astronomical objects. In our previous work, we found that there are 8 physical properties parameters of CLBs located between those of FSRQs and those of BL Lacs. In order to search more CLB candidates (CLBCs), we employed the $mclust$ Gaussian Mixture Modelling clustering algorithm to perform clustering analysis for the 255 subsets of the 8 physical properties parameters with 2250 blazars from the 4FGL-DR3. We find that there are 29 subsets with 3 groups (corresponding to bl lacs, fsrqs, and CLBCs), in which there are 4 subsets with the adjusted Rand index greater then 0.610 (ARI $>$ 0.610). The combined clustering results from 4 subsets report that there are 111 CLBCs that includes 44 CLBs reported in previous literature and 67 new CLBCs, where 11 CLBCs labeled as BL Lac and 56 CLBCs labeled as FSRQ in 4FGL catalog., Comment: Accepted for publication in ApJ. 13 pages, 7 figures, 3 table

This paper proposes a cooperative integrated sensing and communication (ISAC) scheme for the low-altitude sensing scenario, aiming at estimating the parameters of the unmanned aerial vehicles (UAVs) and enhancing the sensing performance via cooperation. The proposed scheme consists of two stages. In Stage I, we formulate the monostatic parameter estimation problem via using a tensor decomposition model. By leveraging the Vandermonde structure of the factor matrix, a spatial smoothing tensor decomposition scheme is introduced to estimate the UAVs' parameters. To further reduce the computational complexity, we design a reduced-dimensional (RD) angle of arrival (AoA) estimation algorithm based on generalized Rayleigh quotient (GRQ). In Stage II, the positions and true velocities of the UAVs are determined through the data fusion across multiple base stations (BSs). Specifically, we first develop a false removing minimum spanning tree (MST)-based data association method to accurately match the BSs' parameter estimations to the same UAV. Then, a Pareto optimality method and a residual weighting scheme are developed to facilitate the position and velocity estimation, respectively. We further extend our approach to the dual-polarized system. Simulation results validate the effectiveness of the proposed schemes in comparison to the conventional techniques.

Using $e^+e^-$ collision data collected with the BESIII detector at the BEPCII collider at thirty-two center-of-mass energies from 3.50 to 4.95 GeV, corresponding to an integrated luminosity of 25 $\rm{fb^{-1}}$, we measure the Born cross section of the $e^+e^-\to\Sigma^0\bar{\Sigma}^0$ reaction and the effective form factor. No significant charmonium(-like) state, i.e., $\psi(3770)$, $\psi(4040)$, $\psi(4160)$, $\psi(4230)$, $\psi(4360)$, $\psi(4415)$, or $\psi(4660)$, decaying into the $\Sigma^0\bar{\Sigma}^0$ final state is observed by fitting the $e^+e^- \to \Sigma^0\bar{\Sigma}^0$ dressed cross section. The upper limits for the product of the branching fraction and the electronic partial width at the 90% confidence level are provided for each assumed charmonium(-like) state. In addition, the ratios of the Born cross section and the effective form factor between the $e^+e^-\to\Sigma^0\bar{\Sigma}^0$ and the $e^+e^-\to\Sigma^+\bar{\Sigma}^-$ reactions are provided, which can be used to validate the prediction of the vector meson dominance model., Comment: 9 pages, 3 figures, 1 Supplemental Material

The rapid development of LLMs has sparked extensive research into their factual knowledge. Current works claim that LLMs fall short on questions requiring less frequent knowledge. However, their proof is incomplete since they only study the influence of entity frequency, which can not fully represent knowledge frequency. So we introduce ComparisonQA benchmark, containing 283K abstract questions, each instantiated by a pair of high-frequency and low-frequency entities. It ensures a controllable comparison because the difference of knowledge frequency between such a pair is only related to entity frequency. In addition, to avoid possible semantic shortcuts, which is a severe problem of current LLMs study, we design a two-round method for knowledge robustness measurement utilizing both correctness and uncertainty. Experiments reveal that LLMs exhibit particularly low robustness regarding low-frequency knowledge, and GPT-4o is even the worst under this measurement. Besides, we introduce an automatic method to filter out questions with low-quality and shortcuts to form ComparisonQA-Hard. We find that uncertainty effectively identifies such questions while maintaining the data size.

Recent advancements in large language models (LLMs) have given rise to the LLM-as-a-judge paradigm, showcasing their potential to deliver human-like judgments. However, in the field of machine translation (MT) evaluation, current LLM-as-a-judge methods fall short of learned automatic metrics. In this paper, we propose Multidimensional Multi-Agent Debate (M-MAD), a systematic LLM-based multi-agent framework for advanced LLM-as-a-judge MT evaluation. Our findings demonstrate that M-MAD achieves significant advancements by (1) decoupling heuristic MQM criteria into distinct evaluation dimensions for fine-grained assessments; (2) employing multi-agent debates to harness the collaborative reasoning capabilities of LLMs; (3) synthesizing dimension-specific results into a final evaluation judgment to ensure robust and reliable outcomes. Comprehensive experiments show that M-MAD not only outperforms all existing LLM-as-a-judge methods but also competes with state-of-the-art reference-based automatic metrics, even when powered by a suboptimal model like GPT-4o mini. Detailed ablations and analysis highlight the superiority of our framework design, offering a fresh perspective for LLM-as-a-judge paradigm. Our code and data are publicly available at https://github.com/SU-JIAYUAN/M-MAD., Comment: Work in progress. Code and data are available at https://github.com/SU-JIAYUAN/M-MAD

The development of the automotive industry and automation has led to a growing demand for time-critical systems to have low latency and jitter for critical traffic. To address this issue, the IEEE 802.1 Time-Sensitive Networking (TSN) task group proposed the Time-Aware Shaper (TAS) to implement Time-Triggered (TT) communication, enabling deterministic transmission by assigning specific time windows to each stream. However, the deployment of TAS requires a solution for the particular deployment scenario. Scheduling algorithms with Fixed-Routing and Waiting-Allowed (FR-WA) mechanisms while providing flexible solutions still have room for optimization in terms of their solution time, reducing network design efficiency and online scheduling deployment. Firstly, the paper overviews the key mechanisms to implement TT communication in TSN, including TAS and FR-WA scheduling algorithms. Secondly, building on this overview of these mechanisms, potential problems with the current implementation of the TAS mechanism are analyzed, including the increasing constraint number as the network scales and potential issues that may arise in traffic anomalies. Then, a method for implementing TT communication using Urgency-Based Scheduler (TT-UBS) is proposed to improve solution efficiency and deterministic transmission in the presence of traffic anomalies. The effectiveness of this method is also analyzed. We propose a scheduling algorithm for solving the parameters of the TT-UBS mechanism. Finally, a simulation-based assessment of the TT-UBS mechanism and the scheduling algorithm is presented. In addition, we extend the method of modifying the scheduling algorithm to other scheduling algorithms and explore the solution efficiency of the extended algorithms.

We present a novel constraint on light dark matter utilizing $1.54$ tonne$\cdot$year of data acquired from the PandaX-4T dual-phase xenon time projection chamber. This constraint is derived through detecting electronic recoil signals resulting from the interaction with solar-enhanced dark matter flux. Low-mass dark matter particles, lighter than a few MeV/$c^2$, can scatter with the thermal electrons in the Sun. Consequently, with higher kinetic energy, the boosted dark matter component becomes detectable via contact scattering with xenon electrons, resulting in a few keV energy deposition that exceeds the threshold of PandaX-4T. We calculate the expected recoil energy in PandaX-4T considering the Sun's acceleration and the detection capabilities of the xenon detector. The first experimental search results using the xenon detector yield the most stringent cross-section of $3.51 \times 10^{-39}~\mathrm{cm}^2$ at $0.08~\mathrm{MeV}$/$c^2$ for a solar boosted dark matter mass ranging from $0.02$ to $10~ \mathrm{MeV}$/$c^2$, achieving a 23 fold improvement compared with earlier experimental studies.

Low-altitude unmanned aerial vehicles (UAVs) are expected to play an important role in future wireless networks, either as aerial base stations (BSs) or aerial users connected to the cellular network. In addition, integrated sensing and communication (ISAC) has been identified as one of the six usage scenarios for the forthcoming sixth-generation (6G) mobile networks, aimed at improving network functionalities and realizing situational awareness of the physical world. While most existing research efforts focus on terrestrial two-dimensional (2D) communication and sensing, UAV as an aerial platform offers a new degree of freedom for designing three-dimensional (3D) air-ground (AG) ISAC networks. In this article, we provide an overview of cellular-connected UAV ISAC, by elaborating the UAV's roles as a target to be sensed and as an aerial anchor to provide sensing functionality, respectively. In particular, we pay attention to the network coverage issue and topics specific to UAV networking, emphasizing the new opportunities as well as unique challenges to be addressed.

Using a data sample of $(10087 \pm 44) \times {10^{6}}$ $J/{\psi}$ events collected with the BESIII detector, we search for the decays $\eta/\eta'\to e^+e^-\mu^+\mu^-$ and $\eta' \to \mu^+\mu^-\mu^+\mu^-$ via the radiative decays $J/{\psi}\to\gamma\eta$/$\gamma\eta'$. No excess of events over expected background is observed for any of the decays of interest. At 90% confidence level, we report the first upper limits on the branching fractions of $\eta' \to e^{+}e^{-}\mu^{+}\mu^{-}$ and $\eta' \to \mu^{+}\mu^{-}\mu^{+}\mu^{-}$ to be $ 1.75 \times {10^{-6}}$ and $5.28 \times {10^{-7}}$, respectively. In addition, we set an upper limit on the branching fraction of $\eta \to e^{+}e^{-}\mu^{+}\mu^{-}$ to be $6.88 \times {10^{-6}}$, which improves the previous result by about two orders of magnitude., Comment: 11 pages

Large-scale superconducting quantum computers require massive numbers of high-performance cryogenic low-noise amplifiers (cryo-LNA) for qubit readout. Here we present a C-Band monolithic microwave integrated circuit (MMIC) cryo-LNA for this purpose. This cryo-LNA is based on 150 nm GaAs pseudomorphic high electron mobility transistor (pHEMT) process and implemented with a three-stage cascaded architecture, where the first stage adopts careful impedance match to optimize the noise and return loss. The integration of negative feedback loops adopted in the second and third-stage enhances the overall stability. Moreover, the pHEMT-self bias and current multiplexing circuitry structure facilitate the reduction of power consumption and require only single bias line. Operating at an ambient temperature of 3.6 K and consuming 15 mW, the cryo-LNA demonstrates good performance in the C-band, reaching a 5 K equivalent noise temperature and an average gain of 40 dB. We further benchmark this cryo-LNA with superconducting qubits, achieving an average single-shot dispersive readout fidelity of 98.3% without assistance from a quantum-limited parametric amplifier. The development of GaAs cryo-LNA diversifies technical support necessary for large-scale quantum applications.

Elastocaloric cooling with shape memory alloys (SMAs) is emerging as a promising candidate for next-generation, environmentally friendly refrigeration. However, its development is hindered by the large driving force and low efficiency associated with uniaxial loading modes. In response, we present an innovative elastocaloric air cooling approach that utilizes the bending of NiTi beams, offering a low-force and energy-efficient solution. We achieve a continuous maximum temperature drop of 0.91 K and a dissipated energy of 15.1 N mm at a low specific driving force of 220 N. Notably, the specimen achieves over 5 million cycles under a maximum surface tensile strain of 1.94% for macroscopic cyclic bending, via a pre-strained, warm laser shock peening (pw-LSP) method. This unprecedented fatigue resistance originates from the formation of a hierarchical microstructure and a large compressive residual stress of over 1 GPa. This work demonstrates the great potential of bending induced elastocaloric cooling in the near future., Comment: 75 pages, 41 figures

Image matching for both cross-view and cross-modality plays a critical role in multimodal perception. In practice, the modality gap caused by different imaging systems/styles poses great challenges to the matching task. Existing works try to extract invariant features for specific modalities and train on limited datasets, showing poor generalization. In this paper, we present MINIMA, a unified image matching framework for multiple cross-modal cases. Without pursuing fancy modules, our MINIMA aims to enhance universal performance from the perspective of data scaling up. For such purpose, we propose a simple yet effective data engine that can freely produce a large dataset containing multiple modalities, rich scenarios, and accurate matching labels. Specifically, we scale up the modalities from cheap but rich RGB-only matching data, by means of generative models. Under this setting, the matching labels and rich diversity of the RGB dataset are well inherited by the generated multimodal data. Benefiting from this, we construct MD-syn, a new comprehensive dataset that fills the data gap for general multimodal image matching. With MD-syn, we can directly train any advanced matching pipeline on randomly selected modality pairs to obtain cross-modal ability. Extensive experiments on in-domain and zero-shot matching tasks, including $19$ cross-modal cases, demonstrate that our MINIMA can significantly outperform the baselines and even surpass modality-specific methods. The dataset and code are available at https://github.com/LSXI7/MINIMA ., Comment: The dataset and code are available at https://github.com/LSXI7/MINIMA

AI-synthesized voice technology has the potential to create realistic human voices for beneficial applications, but it can also be misused for malicious purposes. While existing AI-synthesized voice detection models excel in intra-domain evaluation, they face challenges in generalizing across different domains, potentially becoming obsolete as new voice generators emerge. Current solutions use diverse data and advanced machine learning techniques (e.g., domain-invariant representation, self-supervised learning), but are limited by predefined vocoders and sensitivity to factors like background noise and speaker identity. In this work, we introduce an innovative disentanglement framework aimed at extracting domain-agnostic artifact features related to vocoders. Utilizing these features, we enhance model learning in a flat loss landscape, enabling escape from suboptimal solutions and improving generalization. Extensive experiments on benchmarks show our approach outperforms state-of-the-art methods, achieving up to 5.12% improvement in the equal error rate metric in intra-domain and 7.59% in cross-domain evaluations., Comment: AAAI25

Large language models have already demonstrated their formidable capabilities in general domains, ushering in a revolutionary transformation. However, exploring and exploiting the extensive knowledge of these models to comprehend multi-omics biology remains underexplored. To fill this research gap, we first introduce Biology-Instructions, the first large-scale multi-omics biological sequences-related instruction-tuning dataset including DNA, RNA, proteins, and multi-molecules, designed to bridge the gap between large language models (LLMs) and complex biological sequences-related tasks. This dataset can enhance the versatility of LLMs by integrating diverse biological sequenced-based prediction tasks with advanced reasoning capabilities, while maintaining conversational fluency. Additionally, we reveal significant performance limitations in even state-of-the-art LLMs on biological sequence-related multi-omics tasks without specialized pre-training and instruction-tuning. We further develop a strong baseline called ChatMultiOmics with a novel three-stage training pipeline, demonstrating the powerful ability to understand biology by using Biology-Instructions. Biology-Instructions and ChatMultiOmics are publicly available and crucial resources for enabling more effective integration of LLMs with multi-omics sequence analysis.

Spoken named entity recognition (NER) aims to identify named entities from speech, playing an important role in speech processing. New named entities appear every day, however, annotating their Spoken NER data is costly. In this paper, we demonstrate that existing Spoken NER systems perform poorly when dealing with previously unseen named entities. To tackle this challenge, we propose a method for generating Spoken NER data based on a named entity dictionary (NED) to reduce costs. Specifically, we first use a large language model (LLM) to generate sentences from the sampled named entities and then use a text-to-speech (TTS) system to generate the speech. Furthermore, we introduce a noise metric to filter out noisy data. To evaluate our approach, we release a novel Spoken NER benchmark along with a corresponding NED containing 8,853 entities. Experiment results show that our method achieves state-of-the-art (SOTA) performance in the in-domain, zero-shot domain adaptation, and fully zero-shot settings. Our data will be available at https://github.com/DeepLearnXMU/HeardU., Comment: Accepted by ICASSP 2025

In recent years, anchor and hash-based multi-view clustering methods have gained attention for their efficiency and simplicity in handling large-scale data. However, existing methods often overlook the interactions among multi-view data and higher-order cooperative relationships during projection, negatively impacting the quality of hash representation in low-dimensional spaces, clustering performance, and sensitivity to noise. To address this issue, we propose a novel approach named Tensor-Interacted Projection and Cooperative Hashing for Multi-View Clustering(TPCH). TPCH stacks multiple projection matrices into a tensor, taking into account the synergies and communications during the projection process. By capturing higher-order multi-view information through dual projection and Hamming space, TPCH employs an enhanced tensor nuclear norm to learn more compact and distinguishable hash representations, promoting communication within and between views. Experimental results demonstrate that this refined method significantly outperforms state-of-the-art methods in clustering on five large-scale multi-view datasets. Moreover, in terms of CPU time, TPCH achieves substantial acceleration compared to the most advanced current methods. The code is available at \textcolor{red}{\url{https://github.com/jankin-wang/TPCH}}.

Transferable adversarial examples are known to cause threats in practical, black-box attack scenarios. A notable approach to improving transferability is using integrated gradients (IG), originally developed for model interpretability. In this paper, we find that existing IG-based attacks have limited transferability due to their naive adoption of IG in model interpretability. To address this limitation, we focus on the IG integration path and refine it in three aspects: multiplicity, monotonicity, and diversity, supported by theoretical analyses. We propose the Multiple Monotonic Diversified Integrated Gradients (MuMoDIG) attack, which can generate highly transferable adversarial examples on different CNN and ViT models and defenses. Experiments validate that MuMoDIG outperforms the latest IG-based attack by up to 37.3\% and other state-of-the-art attacks by 8.4\%. In general, our study reveals that migrating established techniques to improve transferability may require non-trivial efforts. Code is available at \url{https://github.com/RYC-98/MuMoDIG}., Comment: Accepted by AAAI 2025

Direct minimization method on the complex Stiefel manifold in Kohn-Sham density functional theory is formulated to treat both finite and extended systems in a unified manner. This formulation is well-suited for scenarios where straightforward iterative diagonalization becomes challenging, especially when the Aufbau principle is not applicable. We present the theoretical foundation and numerical implementation of the Riemannian conjugate gradient (RCG) within a localized non-orthogonal basis set. Riemannian Broyden-Fletcher-Goldfarb-Shanno (RBFGS) method is tentatively implemented. Extensive testing compares the performance of the proposed RCG method with the traditional self-consistent field (SCF) algorithm and shows that it is less efficient. For molecular systems, the RBFGS method requires a computing time comparable to that of SCF calculations. However, for extended systems these methods require much more iterations compared to SCF. Preconditioning can potentially improve its efficiency, especially for metallic systems., Comment: 17 pages, 6 figures

Emotion recognition and touch gesture decoding are crucial for advancing human-robot interaction (HRI), especially in social environments where emotional cues and tactile perception play important roles. However, many humanoid robots, such as Pepper, Nao, and Furhat, lack full-body tactile skin, limiting their ability to engage in touch-based emotional and gesture interactions. In addition, vision-based emotion recognition methods usually face strict GDPR compliance challenges due to the need to collect personal facial data. To address these limitations and avoid privacy issues, this paper studies the potential of using the sounds produced by touching during HRI to recognise tactile gestures and classify emotions along the arousal and valence dimensions. Using a dataset of tactile gestures and emotional interactions from 28 participants with the humanoid robot Pepper, we design an audio-only lightweight touch gesture and emotion recognition model with only 0.24M parameters, 0.94MB model size, and 0.7G FLOPs. Experimental results show that the proposed sound-based touch gesture and emotion recognition model effectively recognises the arousal and valence states of different emotions, as well as various tactile gestures, when the input audio length varies. The proposed model is low-latency and achieves similar results as well-known pretrained audio neural networks (PANNs), but with much smaller FLOPs, parameters, and model size., Comment: ICASSP 2025