Search

Federated Learning (FL) has emerged as a privacy-preserving method for training machine learning models in a distributed manner on edge devices. However, on-device models face inherent computational power and memory limitations, potentially resulting in constrained gradient updates. As the model's size increases, the frequency of gradient updates on edge devices decreases, ultimately leading to suboptimal training outcomes during any particular FL round. This limits the feasibility of deploying advanced and large-scale models on edge devices, hindering the potential for performance enhancements. To address this issue, we propose FedRepOpt, a gradient re-parameterized optimizer for FL. The gradient re-parameterized method allows training a simple local model with a similar performance as a complex model by modifying the optimizer's gradients according to a set of model-specific hyperparameters obtained from the complex models. In this work, we focus on VGG-style and Ghost-style models in the FL environment. Extensive experiments demonstrate that models using FedRepOpt obtain a significant boost in performance of 16.7% and 11.4% compared to the RepGhost-style and RepVGG-style networks, while also demonstrating a faster convergence time of 11.7% and 57.4% compared to their complex structure.

In high temperature cuprate superconductors, it was found that in the overdoped region the superfluid density decreases with the increase of hole doping. One natural question is whether there exists normal fluid in the superconducting state in the overdoped region. In this paper, we have carried out high-resolution ultra-low temperature laser-based angle-resolved photoemission measurements on a heavily overdoped Bi2212 sample with a $T_{\mathrm{c}}$ of 48 K. We find that this heavily overdoped Bi2212 remains in the strong coupling regime with $2 \mathit{\Delta}_0 / k_{\mathrm{B}} T_{\mathrm{c}}=5.8$. The single-particle scattering rate is very small along the nodal direction ($\sim$5 meV) and increases as the momentum moves from the nodal to the antinodal regions. A hard superconducting gap opening is observed near the antinodal region with the spectral weight at the Fermi level fully suppressed to zero. The normal fluid is found to be negligibly small in the superconducting state of this heavily overdoped Bi2212. These results provide key information to understand the high $T_\mathrm{c}$ mechanism in the cuprate superconductors., Comment: 13 pages, 4 figures

Large language models have achieved remarkable success on general NLP tasks, but they may fall short for domain-specific problems. Recently, various Retrieval-Augmented Large Language Models (RALLMs) are proposed to address this shortcoming. However, existing evaluation tools only provide a few baselines and evaluate them on various domains without mining the depth of domain knowledge. In this paper, we address the challenges of evaluating RALLMs by introducing the R-Eval toolkit, a Python toolkit designed to streamline the evaluation of different RAG workflows in conjunction with LLMs. Our toolkit, which supports popular built-in RAG workflows and allows for the incorporation of customized testing data on the specific domain, is designed to be user-friendly, modular, and extensible. We conduct an evaluation of 21 RALLMs across three task levels and two representative domains, revealing significant variations in the effectiveness of RALLMs across different tasks and domains. Our analysis emphasizes the importance of considering both task and domain requirements when choosing a RAG workflow and LLM combination. We are committed to continuously maintaining our platform at https://github.com/THU-KEG/R-Eval to facilitate both the industry and the researchers., Comment: 12 pages, 9 figures, Accepted by KDD2024

This article presents svds-C, an open-source and high-performance C program for accurately and robustly computing truncated SVD, e.g. computing several largest singular values and corresponding singular vectors. We have re-implemented the algorithm of svds in Matlab in C based on MKL or OpenBLAS and multi-thread computing to obtain the parallel program named svds-C. svds-C running on shared-memory computer consumes less time and memory than svds thanks to careful implementation of multi-thread parallelization and memory management. Numerical experiments on different test cases which are synthetically generated or directly from real world datasets show that, svds-C runs remarkably faster than svds with averagely 4.7X and at most 12X speedup for 16-thread parallel computing on a computer with Intel CPU, while preserving same accuracy and consuming about half memory space. Experimental results also demonstrate that svds-C has similar advantages over svds on the computer with AMD CPU, and outperforms other state-of-the-art algorithms for truncated SVD on computing time and robustness., Comment: 20 pages, accepted by SoftwareX

Applying large language models (LLMs) for academic API usage shows promise in reducing researchers' academic information seeking efforts. However, current LLM API-using methods struggle with complex API coupling commonly encountered in academic queries. To address this, we introduce SoAy, a solution-based LLM API-using methodology for academic information seeking. It uses code with a solution as the reasoning method, where a solution is a pre-constructed API calling sequence. The addition of the solution reduces the difficulty for the model to understand the complex relationships between APIs. Code improves the efficiency of reasoning. To evaluate SoAy, we introduce SoAyBench, an evaluation benchmark accompanied by SoAyEval, built upon a cloned environment of APIs from AMiner. Experimental results demonstrate a 34.58-75.99\% performance improvement compared to state-of-the-art LLM API-based baselines. All datasets, codes, tuned models, and deployed online services are publicly accessible at https://github.com/RUCKBReasoning/SoAy., Comment: 22 pages, 13 figures

The Kondo lattice physics, describing the hybridization of localized spin matrix with dispersive conduction electrons, breeds numerous discoveries in the realm of strongly correlated quantum matter. Generally observed in lanthanide and actinide compounds, increasing attention has been directed towards alternative pathways for achieving flat band structures, such as Morie superlattices and Kagome metals. However, fine control of Kondo interaction outside of heterostructures remains elusive. Here we report the discovery of a van der Waals (vdW) kagome antiferromagnet CsCr6Sb6. Angle-resolved photoemission spectra and theoretical analysis show clear flat bands, consisting of half-filled 3dxz and 3dyz orbitals of Cr, situated 50 meV below the Fermi level. Importantly, we observe the emergence of anomalous Hall effect with remarkable tunability by simple reduction the sample thickness. The effective control of kondo interaction in CsCr6Sb6 render it an ideal platform for exploring unpresented phenomena using the vast toolkit of vdW structures.

Aiming to provide a faster and convenient truncated SVD algorithm for large sparse matrices from real applications (i.e. for computing a few of largest singular values and the corresponding singular vectors), a dynamically shifted power iteration technique is applied to improve the accuracy of the randomized SVD method. This results in a dynamic shifts based randomized SVD (dashSVD) algorithm, which also collaborates with the skills for handling sparse matrices. An accuracy-control mechanism is included in the dashSVD algorithm to approximately monitor the per vector error bound of computed singular vectors with negligible overhead. Experiments on real-world data validate that the dashSVD algorithm largely improves the accuracy of randomized SVD algorithm or attains same accuracy with fewer passes over the matrix, and provides an efficient accuracy-control mechanism to the randomized SVD computation, while demonstrating the advantages on runtime and parallel efficiency. A bound of the approximation error of the randomized SVD with the shifted power iteration is also proved., Comment: 26 pages, accepted by ACM Transactions on Mathematical Software

The integration of Federated Learning (FL) and Self-supervised Learning (SSL) offers a unique and synergetic combination to exploit the audio data for general-purpose audio understanding, without compromising user data privacy. However, rare efforts have been made to investigate the SSL models in the FL regime for general-purpose audio understanding, especially when the training data is generated by large-scale heterogeneous audio sources. In this paper, we evaluate the performance of feature-matching and predictive audio-SSL techniques when integrated into large-scale FL settings simulated with non-independently identically distributed (non-iid) data. We propose a novel Federated SSL (F-SSL) framework, dubbed FASSL, that enables learning intermediate feature representations from large-scale decentralized heterogeneous clients, holding unlabelled audio data. Our study has found that audio F-SSL approaches perform on par with the centralized audio-SSL approaches on the audio-retrieval task. Extensive experiments demonstrate the effectiveness and significance of FASSL as it assists in obtaining the optimal global model for state-of-the-art FL aggregation methods.

The latest discovery of high temperature superconductivity near 80K in La3Ni2O7 under high pressure has attracted much attention. Many proposals are put forth to understand the origin of superconductivity.The determination of electronic structures is a prerequisite to establish theories to understand superconductivity in nickelates but is still lacking. Here we report our direct measurement of the electronic structures of La3Ni2O7 by high-resolution angle resolved photoemission spectroscopy. The Fermi surface and band structures of La3Ni2O7 are observed and compared with the band structure calculations. Strong electron correlations are revealed which are orbital- and momentum dependent. A flat band is formed from the Ni-3dz2 orbitals around the zone corner which is ~50meV below the Fermi level and exhibits the strongest electron correlation. In many theoretical proposals, this band is expected to play the dominant role in generating superconductivity in La3Ni2O7. Our observations provide key experimental information to understand the electronic structure and origin of high temperature superconductivity in La3Ni2O7., Comment: 20 pages, 5 figures

This report presents the technical details of our submission to the 2023 Epic-Kitchen EPIC-SOUNDS Audio-Based Interaction Recognition Challenge. The task is to learn the mapping from audio samples to their corresponding action labels. To achieve this goal, we propose a simple yet effective single-stream CNN-based architecture called AudioInceptionNeXt that operates on the time-frequency log-mel-spectrogram of the audio samples. Motivated by the design of the InceptionNeXt, we propose parallel multi-scale depthwise separable convolutional kernels in the AudioInceptionNeXt block, which enable the model to learn the time and frequency information more effectively. The large-scale separable kernels capture the long duration of activities and the global frequency semantic information, while the small-scale separable kernels capture the short duration of activities and local details of frequency information. Our approach achieved 55.43% of top-1 accuracy on the challenge test set, ranked as 1st on the public leaderboard. Codes are available anonymously at https://github.com/StevenLauHKHK/AudioInceptionNeXt.git.

We propose LIGHTNE 2.0, a cost-effective, scalable, automated, and high-quality network embedding system that scales to graphs with hundreds of billions of edges on a single machine. In contrast to the mainstream belief that distributed architecture and GPUs are needed for large-scale network embedding with good quality, we prove that we can achieve higher quality, better scalability, lower cost, and faster runtime with shared-memory, CPU-only architecture. LIGHTNE 2.0 combines two theoretically grounded embedding methods NetSMF and ProNE. We introduce the following techniques to network embedding for the first time: (1) a newly proposed downsampling method to reduce the sample complexity of NetSMF while preserving its theoretical advantages; (2) a high-performance parallel graph processing stack GBBS to achieve high memory efficiency and scalability; (3) sparse parallel hash table to aggregate and maintain the matrix sparsifier in memory; (4) a fast randomized singular value decomposition (SVD) enhanced by power iteration and fast orthonormalization to improve vanilla randomized SVD in terms of both efficiency and effectiveness; (5) Intel MKL for proposed fast randomized SVD and spectral propagation; and (6) a fast and lightweight AutoML library FLAML for automated hyperparameter tuning. Experimental results show that LIGHTNE 2.0 can be up to 84X faster than GraphVite, 30X faster than PBG and 9X faster than NetSMF while delivering better performance. LIGHTNE 2.0 can embed very large graph with 1.7 billion nodes and 124 billion edges in half an hour on a CPU server, while other baselines cannot handle very large graphs of this scale.

A kagome lattice of 3d transition metals hosts flat bands, Dirac fermions and saddle points. It provides a versatile platform for achieving topological superconductivity, anomalous Hall effect, unconventional density wave order and quantum spin liquid when the strong correlation, spin-orbit coupling or magnetic order are involved in such a lattice. Here, using laser-based angle-resolved photoemission spectroscopy in combination with density functional theory calculations, we investigate the electronic structure of the newly discovered kagome superconductor CsTi$_3$Bi$_5$, which is isostructural to the AV$_3$Sb$_5$ (A=K, Rb or Cs) kagome superconductors and possesses a perfect two-dimensional kagome network of Titanium. We directly observed a strikingly flat band derived from the local destructive interferences of Bloch wave functions within the kagome lattices. We also identify the type-II Dirac nodal loops around the Brillouin zone center, the type-III Dirac nodal loops around the zone corners and type-III Dirac nodal lines along the k$_z$ direction. In addition, around the Brillouin zone center, Z2 nontrivial topological surface states are also observed which is formed from the band inversion due to strong spin orbital coupling. The simultaneous existence of such multi-sets of nontrivial band structures in one kagome superconductor not only provides good opportunities to study related physics in the kagome lattice but also makes CsTi$_3$Bi$_5$ an ideal system to realize noval quantum phenomena by manipulating its chemical potential with chemical doping or pressure.

Given a stream of money transactions between accounts in a bank, how can we accurately detect money laundering agent accounts and suspected behaviors in real-time? Money laundering agents try to hide the origin of illegally obtained money by dispersive multiple small transactions and evade detection by smart strategies. Therefore, it is challenging to accurately catch such fraudsters in an unsupervised manner. Existing approaches do not consider the characteristics of those agent accounts and are not suitable to the streaming settings. Therefore, we propose MonLAD and MonLAD-W to detect money laundering agent accounts in a transaction stream by keeping track of their residuals and other features; we devise AnoScore algorithm to find anomalies based on the robust measure of statistical deviation. Experimental results show that MonLAD outperforms the state-of-the-art baselines on real-world data and finds various suspicious behavior patterns of money laundering. Additionally, several detected suspected accounts have been manually-verified as agents in real money laundering scenario.

We study large-scale network embedding with the goal of generating high-quality embeddings for networks with more than 1 billion vertices and 100 billion edges. Recent attempts LightNE and NetSMF propose to sparsify and factorize the (dense) NetMF matrix for embedding large networks, where NetMF is a theoretically-grounded network embedding method. However, there is a trade-off between their embeddings' quality and scalability due to their expensive memory requirements, making embeddings less effective under real-world memory constraints. Therefore, we present the SketchNE model, a scalable, effective, and memory-efficient network embedding solution developed for a single machine with CPU only. The main idea of SketchNE is to avoid the explicit construction and factorization of the NetMF matrix either sparsely or densely when producing the embeddings through the proposed sparse-sign randomized single-pass SVD algorithm. We conduct extensive experiments on nine datasets of various sizes for vertex classification and link prediction, demonstrating the consistent outperformance of SketchNE over state-of-the-art baselines in terms of both effectiveness and efficiency. SketchNE costs only 1.0 hours to embed the Hyperlink2012 network with 3.5 billion vertices and 225 billion edges on a CPU-only single machine with embedding superiority (e.g., a 282% relative HITS@10 gain over LightNE).

A model-based collaborative filtering (CF) approach utilizing fast adaptive randomized singular value decomposition (SVD) is proposed for the matrix completion problem in recommender system. Firstly, a fast adaptive PCA frameworkis presented which combines the fixed-precision randomized matrix factorization algorithm [1] and accelerating skills for handling large sparse data. Then, a novel termination mechanism for the adaptive PCA is proposed to automatically determine a number of latent factors for achieving the near optimal prediction accuracy during the subsequent model-based CF. The resulted CF approach has good accuracy while inheriting high runtime efficiency. Experiments on real data show that, the proposed adaptive PCA is up to 2.7X and 6.7X faster than the original fixed-precision SVD approach [1] and svds in Matlab repsectively, while preserving accuracy. The proposed model-based CF approach is able to efficiently process the MovieLens data with 20M ratings and exhibits more than 10X speedup over the regularized matrix factorization based approach [2] and the fast singular value thresholding approach [3] with comparable or better accuracy. It also owns the advantage of parameter free. Compared with the deep-learning-based CF approach, the proposed approach is much more computationally efficient, with just marginal performance loss.