arXiv subjects

Can Wang

Publications and source records attributed to Can Wang.

At least 109 records · Page 6Linked to original sources

arXiv2025

Bridging the Gap: Self-Optimized Fine-Tuning for LLM-based Recommender Systems

Recent years have witnessed extensive exploration of Large Language Models (LLMs) on the field of Recommender Systems (RS). There are currently two commonly used strategies to enable LLMs to have recommendation capabilities: 1) The "Guidance-Only" strategy uses in-context learning to exploit and amplify the inherent semantic understanding and item recommendation capabilities of LLMs; 2) The "Tuning-Only" strategy uses supervised fine-tuning (SFT) to fine-tune LLMs with the aim of fitting them to real recommendation data. However, neither of these strategies can effectively bridge the gap between the knowledge space of LLMs and recommendation, and their performance do not meet our expectations. To better enable LLMs to learn recommendation knowledge, we combine the advantages of the above two strategies and proposed a novel "Guidance+Tuning" method called Self-Optimized Fine-Tuning (SOFT), which adopts the idea of curriculum learning. It first employs self-distillation to construct an auxiliary easy-to-learn but meaningful dataset from a fine-tuned LLM. Then it further utilizes a self-adaptive curriculum scheduler to enable LLMs to gradually learn from simpler data (self-distilled data) to more challenging data (real RS data). Extensive experiments demonstrate that SOFT significantly enhances the recommendation accuracy (37.59\% on average) of LLM-based methods. The code is available via https://anonymous.4open.science/r/Self-Optimized-Fine-Tuning-264E

cs.IR↗

arXiv2025

Advancing Antiferromagnetic Nitrides via Metal Alloy Nitridation

Nitride materials, valued for their structural stability and exceptional physical properties, have garnered significant interest in both fundamental research and technological applications. The fabrication of high-quality nitride thin films is essential for advancing their use in microelectronics and spintronics. Yet, achieving single-crystal nitride thin films with excellent structural integrity remains a challenge. Here, we introduce a straightforward yet innovative metallic alloy nitridation technique for the synthesis of stable single-crystal nitride thin films. By subjecting metal alloy thin films to a controlled nitridation process, nitrogen atoms integrate into the lattice, driving structural transformations while preserving high epitaxial quality. Combining nanoscale magnetic imaging with a diamond nitrogen-vacancy (NV) probe, X-ray magnetic linear dichroism, and comprehensive transport measurements, we confirm that the nitridated films exhibit a robust antiferromagnetic character with a zero net magnetic moment. This work not only provides a refined and reproducible strategy for the fabrication of nitride thin films but also lays a robust foundation for exploring their burgeoning device applications.

cond-mat.mtrl-sci↗

arXiv2025

Room-temperature magnetoelectric coupling in strontium titanate

Magnetoelectric (ME) multiferroics enable efficient interconversion of electrical and magnetic signals, offering pathways toward substantial reduction of power consumption in next-generation computing and information technologies. However, despite decades of research, the persistence of ME coupling at room-temperature, an indispensable feature for practical applications, remains exceedingly rare in single-phase materials, due to symmetry constraints imposed by magnetic point groups and competing electronic requirements for ferroelectricity and magnetism. Here we report the coexistence of ferroelectricity and ferromagnetism at 780 K and a strong ME coupling with a converse ME coupling coefficient up to 498 ps/m at room-temperature in strontium titanate by dedicated vacancy engineering. The asymmetric distribution of oxygen vacancies surrounding titanium vacancies enables atomic displacement of titanium and charge injection, providing joint origin for ferroelectricity and ferromagnetism, as confirmed by atomic-scale structural analysis and first-principles calculations. The mechanism of ME coupling is offered as the hopping of oxygen vacancies under electric fields, which leads to the rearrangement of electronic configuration. Our work opens an avenue for designing multiferroics, overcoming the long-standing scarcity of room-temperature single-phase multiferroics.

cond-mat.mtrl-sci↗

arXiv2025

A single-phase epitaxially grown ferroelectric perovskite nitride

The integration of ferroelectrics with semiconductors is crucial for developing functional devices, such as field-effect transistors, tunnel junctions, and nonvolatile memories. However, the synthesis of high-quality single-crystalline ferroelectric nitride perovskites has been limited, hindering a comprehensive understanding of their switching dynamics and potential applications. Here we report the synthesis and characterizations of epitaxial single-phase ferroelectric cerium tantalum nitride (CeTaN3) on both oxides and semiconductors. The polar symmetry of CeTaN3 was confirmed by observing the atomic displacement of central ions relative to the center of the TaN6 octahedra, as well as through optical second harmonic generation. We observed switchable ferroelectric domains in CeTaN3 films using piezo-response force microscopy, complemented by the characterization of square-like polarization-electric field hysteresis loops. The remanent polarization of CeTaN3 reaches approximately 20 uC/cm2 at room temperature, consistent with theoretical calculations. This work establishes a vital link between ferroelectric nitride perovskites and their practical applications, paving the way for next-generation information and energy-storage devices with enhanced performance, scalability, and manufacturability.

cond-mat.mtrl-sci↗

arXiv2025

Deteriorated Interlayer Coupling in Twisted Bilayer Cobaltites

A wealth of remarkable behaviors is observed at the interfaces between magnetic oxides due to the coexistence of Coulomb repulsion and interatomic exchange interactions. While previous research has focused on bonded oxide heterointerfaces, studies on magnetism in van der Waals interfaces remain rare. In this study, we stacked two freestanding cobaltites with precisely controlled twist angles. Scanning transmission electron microscopy revealed clear and ordered moiré patterns, which exhibit an inverse relationship with the twist angle. We found that the Curie temperature in the twisted region is reduced by approximately 13 K compared to the single-layer region using nitrogen-vacancy (NV) magnetometry. This phenomenon may be related to the weakening of the orbital hybridization between oxygen ions and transition metal ions in the unbonded interfaces. Our findings suggest a potential avenue for modulating magnetic interactions in correlated systems through twist, providing opportunities for the discovery of unknown quantum states.

cond-mat.mtrl-sci↗

arXiv2025

Charge transfer induced insulating state at antiperovskite/perovskite heterointerfaces

Heterointerfaces have been pivotal in unveiling extraordinary interfacial properties and enabling multifunctional material platforms. Despite extensive research on all-oxide interfaces, heterointerfaces between different material classes, such as oxides and nitrides, remain underexplored. Here we present the fabrication of a high-quality Dirac metal antiperovskite Ni3InN, characterized by an extremely low temperature coefficient of resistivity, approximately 1.8*10^-8 Ω*cm/K, over a broad temperature range. Atomically sharp heterointerfaces between Ni3InN and SrVO3 were constructed, revealing intriguing interfacial phenomena. Leveraging layer-resolved scanning transmission electron microscopy and electron energy loss spectroscopy, we identified pronounced charge transfer across the well-ordered interface. Remarkably, this interfacial electron transfer from Ni3InN to SrVO3 induces an insulating interfacial layer and an emergent magnetic moment within the Ni3InN layer, consistent with first-principles calculations. These findings pave the way for novel electronic and spintronic applications by enabling tunable interfacial properties in nitride/oxide systems.

cond-mat.mtrl-sci↗

arXiv2025

How Do Recommendation Models Amplify Popularity Bias? An Analysis from the Spectral Perspective

Recommendation Systems (RS) are often plagued by popularity bias. When training a recommendation model on a typically long-tailed dataset, the model tends to not only inherit this bias but often exacerbate it, resulting in over-representation of popular items in the recommendation lists. This study conducts comprehensive empirical and theoretical analyses to expose the root causes of this phenomenon, yielding two core insights: 1) Item popularity is memorized in the principal spectrum of the score matrix predicted by the recommendation model; 2) The dimension collapse phenomenon amplifies the relative prominence of the principal spectrum, thereby intensifying the popularity bias. Building on these insights, we propose a novel debiasing strategy that leverages a spectral norm regularizer to penalize the magnitude of the principal singular value. We have developed an efficient algorithm to expedite the calculation of the spectral norm by exploiting the spectral property of the score matrix. Extensive experiments across seven real-world datasets and three testing paradigms have been conducted to validate the superiority of the proposed method.

cs.IR↗

arXiv2025

Generative Object Insertion in Gaussian Splatting with a Multi-View Diffusion Model

Generating and inserting new objects into 3D content is a compelling approach for achieving versatile scene recreation. Existing methods, which rely on SDS optimization or single-view inpainting, often struggle to produce high-quality results. To address this, we propose a novel method for object insertion in 3D content represented by Gaussian Splatting. Our approach introduces a multi-view diffusion model, dubbed MVInpainter, which is built upon a pre-trained stable video diffusion model to facilitate view-consistent object inpainting. Within MVInpainter, we incorporate a ControlNet-based conditional injection module to enable controlled and more predictable multi-view generation. After generating the multi-view inpainted results, we further propose a mask-aware 3D reconstruction technique to refine Gaussian Splatting reconstruction from these sparse inpainted views. By leveraging these fabricate techniques, our approach yields diverse results, ensures view-consistent and harmonious insertions, and produces better object quality. Extensive experiments demonstrate that our approach outperforms existing methods.

cs.CV↗

arXiv2025

Rankformer: A Graph Transformer for Recommendation based on Ranking Objective

Recommender Systems (RS) aim to generate personalized ranked lists for each user and are evaluated using ranking metrics. Although personalized ranking is a fundamental aspect of RS, this critical property is often overlooked in the design of model architectures. To address this issue, we propose Rankformer, a ranking-inspired recommendation model. The architecture of Rankformer is inspired by the gradient of the ranking objective, embodying a unique (graph) transformer architecture -- it leverages global information from all users and items to produce more informative representations and employs specific attention weights to guide the evolution of embeddings towards improved ranking performance. We further develop an acceleration algorithm for Rankformer, reducing its complexity to a linear level with respect to the number of positive instances. Extensive experimental results demonstrate that Rankformer outperforms state-of-the-art methods. The code is available at https://github.com/StupidThree/Rankformer.

cs.IR↗

arXiv2025

Full Polarization Control of Photons with Evanescent Wave Coupling in the Ultra Subwavelength Gap of Photonic Molecules

Polarization of photons plays a key role in quantum optics and light-matter interactions, however, it is difficult to control in nanosystems since the eigenstate of a nanophotonic cavity is usually fixed and linearly polarized. Here we reveal polarization control of photons using photonic molecules (PMs) that host supermodes of two coupled nanobeam cavities. In contrast to conventional PMs in a 2D photonic crystal slab, for the two 1D photonic crystal nanobeam cavities the shift and gap between them can be tuned continuously. With an ultra subwavelength gap, the coupling between the two cavities is dominated by the evanescent wave coupling in the surrounding environment, rather not the emission wave coupling for conventional PMs. As such, non-Hermiticity of the system becomes pronounced, and the supermodes consist of a non-trivial phase difference between bare eigenstates that supports elliptical polarization. We observe that both the polarization degree and polarization angle of the antisymmetric mode strongly depend on the shift and gap between the two cavities, exhibiting polarization states from linear to circular. This full polarization control indicates great potential of PMs in quantum optical devices and spin-resolved cavity quantum electrodynamics.

physics.optics↗

arXiv2025

Trunk-branch Contrastive Network with Multi-view Deformable Aggregation for Multi-view Action Recognition

Multi-view action recognition aims to identify actions in a given multi-view scene. Traditional studies initially extracted refined features from each view, followed by implemented paired interaction and integration, but they potentially overlooked the critical local features in each view. When observing objects from multiple perspectives, individuals typically form a comprehensive impression and subsequently fill in specific details. Drawing inspiration from this cognitive process, we propose a novel trunk-branch contrastive network (TBCNet) for RGB-based multi-view action recognition. Distinctively, TBCNet first obtains fused features in the trunk block and then implicitly supplements vital details provided by the branch block via contrastive learning, generating a more informative and comprehensive action representation. Within this framework, we construct two core components: the multi-view deformable aggregation and the trunk-branch contrastive learning. MVDA employed in the trunk block effectively facilitates multi-view feature fusion and adaptive cross-view spatio-temporal correlation, where a global aggregation module is utilized to emphasize significant spatial information and a composite relative position bias is designed to capture the intra- and cross-view relative positions. Moreover, a trunk-branch contrastive loss is constructed between aggregated features and refined details from each view. By incorporating two distinct weights for positive and negative samples, a weighted trunk-branch contrastive loss is proposed to extract valuable information and emphasize subtle inter-class differences. The effectiveness of TBCNet is verified by extensive experiments on four datasets including NTU-RGB+D 60, NTU-RGB+D 120, PKU-MMD, and N-UCLA dataset. Compared to other RGB-based methods, our approach achieves state-of-the-art performance in cross-subject and cross-setting protocols.

cs.CV↗

arXiv2025

Uncertainty-Aware Graph Structure Learning

Graph Neural Networks (GNNs) have become a prominent approach for learning from graph-structured data. However, their effectiveness can be significantly compromised when the graph structure is suboptimal. To address this issue, Graph Structure Learning (GSL) has emerged as a promising technique that refines node connections adaptively. Nevertheless, we identify two key limitations in existing GSL methods: 1) Most methods primarily focus on node similarity to construct relationships, while overlooking the quality of node information. Blindly connecting low-quality nodes and aggregating their ambiguous information can degrade the performance of other nodes. 2) The constructed graph structures are often constrained to be symmetric, which may limit the model's flexibility and effectiveness. To overcome these limitations, we propose an Uncertainty-aware Graph Structure Learning (UnGSL) strategy. UnGSL estimates the uncertainty of node information and utilizes it to adjust the strength of directional connections, where the influence of nodes with high uncertainty is adaptively reduced. Importantly, UnGSL serves as a plug-in module that can be seamlessly integrated into existing GSL methods with minimal additional computational cost. In our experiments, we implement UnGSL into six representative GSL methods, demonstrating consistent performance improvements.

cs.LG↗

arXiv2025

sEMG-Based Joint Angle Estimation via Hierarchical Spiking Attentional Feature Decomposition Network

Surface electromyography (sEMG) has demonstrated significant potential in simultaneous and proportional control (SPC). However, existing algorithms for predicting joint angles based onsEMGoften suffer fromhigh inference costs or are limited to specific subjects rather than multi-subject scenarios. To address these challenges, we introduced a hierarchical Spiking Attentional Feature Decomposition Network (SAFE-Net). This network initially compresses sEMG signals into neural spiking forms using a Spiking Sparse Attention Encoder (SSAE). Subsequently, the compressed features are decomposed into kinematic and biological features through a Spiking Attentional Feature Decomposition (SAFD) module. Finally, the kinematic and biological features are used to predict joint angles and identify subject identities, respectively. Our validation on two datasets and comparison with two existing methods, Informer and Spikformer, demonstrate that SSAE achieves significant power consumption savings of 39.1% and 37.5% respectively over them in terms of inference costs. Furthermore, SAFE-Net surpasses Informer and Spikformer in recognition accuracy on both datasets. This study underscores the potential of SAFE-Net to advance the field of SPC in lower limb rehabilitation exoskeleton robots.

cs.HC↗

arXiv2024

A Framework for Effective Invocation Methods of Various LLM Services

Large Language Models (LLMs) have shown impressive abilities in solving various natural language processing tasks and are now widely offered as services. LLM services enable users to accomplish tasks without requiring specialized knowledge, simply by paying service providers. However, numerous providers offer various LLM services with variations in pricing, latency, and performance. These factors are also affected by different invocation methods, such as the choice of context and the use of cache, which lead to unpredictable and uncontrollable service cost and quality. Consequently, utilizing various LLM services invocation methods to construct an effective (cost-saving, low-latency and high-performance) invocation strategy that best meets task demands becomes a pressing challenge. This paper provides a comprehensive overview of methods help LLM services to be invoked efficiently. Technically, we define the problem of constructing an effective LLM services invocation strategy, and based on this, propose a unified LLM service invocation framework. The framework classifies existing methods into four categories: input abstraction, semantic cache, solution design, and output enhancement, which can be used separately or jointly during the invocation life cycle. We discuss the methods in each category and compare them to provide valuable guidance for researchers. Finally, we emphasize the open challenges in this domain and shed light on future research.

cs.SE↗

arXiv2024

99.9%-fidelity in measuring a superconducting qubit

Despite the significant progress in superconducting quantum computation over the past years, quantum state measurement still lags nearly an order of magnitude behind quantum gate operations in speed and fidelity. The main challenge is that the strong coupling and readout signal used to probe the quantum state may also introduce additional channels which may cause qubit state transitions. Here, we design a novel architecture to implement the long-sought longitudinal interaction scheme between qubits and resonators. This architecture not only provides genuine longitudinal interaction by eliminating residual transversal couplings, but also introduces proper nonlinearity to the resonator that can further minimize decay error and measurement-induced excitation error. Our experimental results demonstrate a measurement fidelity of 99.8% in 202 ns without the need for any first-stage amplification. After subtracting the residual preparation errors, the pure measurement fidelity is above 99.9%. Our scheme is compatible with the multiplexing readout scheme and can be used for quantum error correction.

quant-ph↗

arXiv2024

Online Adversarial Knowledge Distillation for Graph Neural Networks

Knowledge distillation, a technique recently gaining popularity for enhancing model generalization in Convolutional Neural Networks (CNNs), operates under the assumption that both teacher and student models are trained on identical data distributions. However, its effect on Graph Neural Networks (GNNs) is less than satisfactory since the graph topology and node attributes are prone to evolve, thereby leading to the issue of distribution shift. In this paper, we tackle this challenge by simultaneously training a group of graph neural networks in an online distillation fashion, where the group knowledge plays a role as a dynamic virtual teacher and the structure changes in graph neural networks are effectively captured. To improve the distillation performance, two types of knowledge are transferred among the students to enhance each other: local knowledge reflecting information in the graph topology and node attributes, and global knowledge reflecting the prediction over classes. We transfer the global knowledge with KL-divergence as the vanilla knowledge distillation does, while exploiting the complicated structure of the local knowledge with an efficient adversarial cyclic learning framework. Extensive experiments verified the effectiveness of our proposed online adversarial distillation approach. The code is published at https://github.com/wangz3066/OnlineDistillGCN.

cs.LG↗

arXiv2024

Dynamic Graph Transformer with Correlated Spatial-Temporal Positional Encoding

Learning effective representations for Continuous-Time Dynamic Graphs (CTDGs) has garnered significant research interest, largely due to its powerful capabilities in modeling complex interactions between nodes. A fundamental and crucial requirement for representation learning in CTDGs is the appropriate estimation and preservation of proximity. However, due to the sparse and evolving characteristics of CTDGs, the spatial-temporal properties inherent in high-order proximity remain largely unexplored. Despite its importance, this property presents significant challenges due to the computationally intensive nature of personalized interaction intensity estimation and the dynamic attributes of CTDGs. To this end, we propose a novel Correlated Spatial-Temporal Positional encoding that incorporates a parameter-free personalized interaction intensity estimation under the weak assumption of the Poisson Point Process. Building on this, we introduce the Dynamic Graph Transformer with Correlated Spatial-Temporal Positional Encoding (CorDGT), which efficiently retains the evolving spatial-temporal high-order proximity for effective node representation learning in CTDGs. Extensive experiments on seven small and two large-scale datasets demonstrate the superior performance and scalability of the proposed CorDGT. The code is available at: https://github.com/wangz3066/CorDGT.

cs.LG↗

arXiv2024

PSL: Rethinking and Improving Softmax Loss from Pairwise Perspective for Recommendation

Softmax Loss (SL) is widely applied in recommender systems (RS) and has demonstrated effectiveness. This work analyzes SL from a pairwise perspective, revealing two significant limitations: 1) the relationship between SL and conventional ranking metrics like DCG is not sufficiently tight; 2) SL is highly sensitive to false negative instances. Our analysis indicates that these limitations are primarily due to the use of the exponential function. To address these issues, this work extends SL to a new family of loss functions, termed Pairwise Softmax Loss (PSL), which replaces the exponential function in SL with other appropriate activation functions. While the revision is minimal, we highlight three merits of PSL: 1) it serves as a tighter surrogate for DCG with suitable activation functions; 2) it better balances data contributions; and 3) it acts as a specific BPR loss enhanced by Distributionally Robust Optimization (DRO). We further validate the effectiveness and robustness of PSL through empirical experiments. The code is available at https://github.com/Tiny-Snow/IR-Benchmark.

cs.LG↗