arXiv subjects

Jun Zhang

Publications and source records attributed to Jun Zhang.

At least 433 records · Page 24Linked to original sources

arXiv2025

CAB: Comprehensive Attention Benchmarking on Long Sequence Modeling

Transformer has achieved remarkable success in language, image, and speech processing. Recently, various efficient attention architectures have been proposed to improve transformer's efficiency while largely preserving its efficacy, especially in modeling long sequences. A widely-used benchmark to test these efficient methods' capability on long-range modeling is Long Range Arena (LRA). However, LRA only focuses on the standard bidirectional (or noncausal) self attention, and completely ignores cross attentions and unidirectional (or causal) attentions, which are equally important to downstream applications. In this paper, we propose Comprehensive Attention Benchmark (CAB) under a fine-grained attention taxonomy with four distinguishable attention patterns, namely, noncausal self, causal self, noncausal cross, and causal cross attentions. CAB collects seven real-world tasks from different research areas to evaluate efficient attentions under the four attention patterns. Among these tasks, CAB validates efficient attentions in eight backbone networks to show their generalization across neural architectures. We conduct exhaustive experiments to benchmark the performances of nine widely-used efficient attention architectures designed with different philosophies on CAB. Extensive experimental results also shed light on the fundamental problems of efficient attentions, such as efficiency length against vanilla attention, performance consistency across attention patterns, the benefit of attention mechanisms, and interpolation/extrapolation on long-context language modeling.

cs.LG↗

arXiv2025

Emergence of the Traffic Autonomous Zone (TAZ) for Telecommunication Operations from Spatial Heterogeneity in Cellular Networks

In the field of telecommunications, various operations are driven by different physical quantities. Each has its own patterns in time and space, but all show some clustered structures in their spatial distribution. This reflects a unified rule of human mobility, suggesting the consistency among different telecommunication regionalization objectives. With this in mind, regionalization can be used to identify these patterns and can be applied to improve management efficiency in the context of "autonomous networks". This article introduces the "Traffic Autonomous Zone (TAZ)" concept. This approach aims to create a reasonable unified regionalization scheme by identifying spatial clusters. It is not just a practical way to partition cities based on telecommunications needs, but it also captures self-organization structure of cities in essence. We present examples of this regionalization method using real data. Compared to the popular Louvain community detection method, our approach is on the Pareto frontier, allowing for a balance among various metrics in telecommunications.

cs.SI↗

arXiv2025

Dual Mutual Learning Network with Global-local Awareness for RGB-D Salient Object Detection

RGB-D salient object detection (SOD), aiming to highlight prominent regions of a given scene by jointly modeling RGB and depth information, is one of the challenging pixel-level prediction tasks. Recently, the dual-attention mechanism has been devoted to this area due to its ability to strengthen the detection process. However, most existing methods directly fuse attentional cross-modality features under a manual-mandatory fusion paradigm without considering the inherent discrepancy between the RGB and depth, which may lead to a reduction in performance. Moreover, the long-range dependencies derived from global and local information make it difficult to leverage a unified efficient fusion strategy. Hence, in this paper, we propose the GL-DMNet, a novel dual mutual learning network with global-local awareness. Specifically, we present a position mutual fusion module and a channel mutual fusion module to exploit the interdependencies among different modalities in spatial and channel dimensions. Besides, we adopt an efficient decoder based on cascade transformer-infused reconstruction to integrate multi-level fusion features jointly. Extensive experiments on six benchmark datasets demonstrate that our proposed GL-DMNet performs better than 24 RGB-D SOD methods, achieving an average improvement of ~3% across four evaluation metrics compared to the second-best model (S3Net). Codes and results are available at https://github.com/kingkung2016/GL-DMNet.

cs.CV↗

arXiv2024

One Pot Synthesis of Cubic Gauche Polymeric Nitrogen

The long sought cubic gauche polymeric nitrogen (cg-N) consisting of N-N single bonds has been synthesized by a simple route using sodium azide as a precursor at ambient conditions. The recrystallization process was designed to expose crystal faces with low activation energy that facilitates initiating the polymeric reaction at ambient conditions. The azide was considered as a precursor due to the low energy barrier in transforming double bonded N=N to single bonded cg-N. Raman spectrum measurements detected the emerging vibron peaks at 635 cm-1 for the polymerized sodium azide samples, demonstrating the formation of cg-N with N-N single bonds. Different from traditional high pressure technique and recently developed plasma enhanced chemical vapor deposition method, the route achieves the quantitative synthesis of cg-N at ambient conditions. The simple method to synthesize cg-N offers potential for further scale up production as well as practical applications of polymeric nitrogen based materials as high energy density materials.

cond-mat.mtrl-sci↗

arXiv2024

Level-Navi Agent: A Framework and benchmark for Chinese Web Search Agents

Large language models (LLMs), adopted to understand human language, drive the development of artificial intelligence (AI) web search agents. Compared to traditional search engines, LLM-powered AI search agents are capable of understanding and responding to complex queries with greater depth, enabling more accurate operations and better context recognition. However, little attention and effort has been paid to the Chinese web search, which results in that the capabilities of open-source models have not been uniformly and fairly evaluated. The difficulty lies in lacking three aspects: an unified agent framework, an accurately labeled dataset, and a suitable evaluation metric. To address these issues, we propose a general-purpose and training-free web search agent by level-aware navigation, Level-Navi Agent, accompanied by a well-annotated dataset (Web24) and a suitable evaluation metric. Level-Navi Agent can think through complex user questions and conduct searches across various levels on the internet to gather information for questions. Meanwhile, we provide a comprehensive evaluation of state-of-the-art LLMs under fair settings. To further facilitate future research, source code is available at Github.

cs.IR↗

arXiv2024

A Rb-Cs dual-species magneto-optical trap

We describe a three-dimensional (3D) magneto-optical trap (MOT) capable of simultaneously capturing 85Rb and 133Cs atoms. Unlike conventional setups, our system utilizes two separate laser systems that are combined before entering the vacuum chamber, enabling the simultaneous trapping of two different atomic species. Additionally, in our 3D MOT configuration, two (of three) pairs of laser beams are not orthogonal to the chamber surfaces but are aligned at a 45° angle. With a total trapping laser power of 8 mW and repump laser power of 4 mW for Rb atoms, and a total trapping laser power of 7.5 mW and repump laser power of 1.5 mW for Cs atoms, we achieve optical depths (OD) of 3.71 for Rb and 3.45 for Cs, demonstrating efficient trapping for both species. Our 3D MOT setup allows full horizontal optical access to the trapped atomic ensembles without spatial interference from the trapping or repump laser beams. Moreover, the red detuning for trapping both atomic species is smaller than in traditional configurations. This system offers a versatile platform for exploring complex phenomena in ultracold atom physics, such as Rydberg molecule formation and interspecies interactions.

physics.atom-ph↗

arXiv2024

Deep Learning Models for Colloidal Nanocrystal Synthesis

Colloidal synthesis of nanocrystals usually includes complex chemical reactions and multi-step crystallization processes. Despite the great success in the past 30 years, it remains challenging to clarify the correlations between synthetic parameters of chemical reaction and physical properties of nanocrystals. Here, we developed a deep learning-based nanocrystal synthesis model that correlates synthetic parameters with the final size and shape of target nanocrystals, using a dataset of 3500 recipes covering 348 distinct nanocrystal compositions. The size and shape labels were obtained from transmission electron microscope images using a segmentation model trained with a semi-supervised algorithm on a dataset comprising 1.2 million nanocrystals. By applying the reaction intermediate-based data augmentation method and elaborated descriptors, the synthesis model was able to predict nanocrystal's size with a mean absolute error of 1.39 nm, while reaching an 89% average accuracy for shape classification. The synthesis model shows knowledge transfer capabilities across different nanocrystals with inputs of new recipes. With that, the influence of chemicals on the final size of nanocrystals was further evaluated, revealing the importance order of nanocrystal composition, precursor or ligand, and solvent. Overall, the deep learning-based nanocrystal synthesis model offers a powerful tool to expedite the development of high-quality nanocrystals.

cond-mat.mtrl-sci↗

arXiv2024

Learn How to Query from Unlabeled Data Streams in Federated Learning

Federated learning (FL) enables collaborative learning among decentralized clients while safeguarding the privacy of their local data. Existing studies on FL typically assume offline labeled data available at each client when the training starts. Nevertheless, the training data in practice often arrive at clients in a streaming fashion without ground-truth labels. Given the expensive annotation cost, it is critical to identify a subset of informative samples for labeling on clients. However, selecting samples locally while accommodating the global training objective presents a challenge unique to FL. In this work, we tackle this conundrum by framing the data querying process in FL as a collaborative decentralized decision-making problem and proposing an effective solution named LeaDQ, which leverages multi-agent reinforcement learning algorithms. In particular, under the implicit guidance from global information, LeaDQ effectively learns the local policies for distributed clients and steers them towards selecting samples that can enhance the global model's accuracy. Extensive simulations on image and text tasks show that LeaDQ advances the model performance in various FL scenarios, outperforming the benchmarking algorithms.

cs.LG↗

arXiv2024

The re-entrant ferromagnetism due to symmetric exchange bias in two septuple-layer MnBi2Te4 epitaxial films

MnBi2Te4 (MBT) is a typical magnetic topological insulator with an A-type antiferromagnetic (AFM) ground state. Here we prepared ultra-thin MBT films with controlled anti-site defects and observed rich doping-dependent magnetic behaviors. We find in one-septuple-layer MBT films a ferrimagnetic ground state and in two-septuple-layer ones a kind of re-entrant ferromagnetism (FM) that disappears with increased Bi-on-Mn doping in the FM Mn-layers. This re-entrant behavior is attributed to a kind of symmetric exchange-bias effect that arises in the presence of both AFM and FM sub-systems due to the introduction of high-dense Mn-on-Bi anti-site defects. Furthermore, all MBT films display spin-glass-like behaviors. Our work demonstrates rich magnetic behaviors originating from the competing magnetic interactions between Mn spins at different lattice positions in MBT.

cond-mat.mtrl-sci↗

arXiv2024

Folded multistability and hidden critical point in microwave-driven Rydberg atoms

The interactions between Rydberg atoms and microwave fields provide a valuable framework for studying the complex dynamics out of equilibrium, exotic phases, and critical phenomena in many-body physics. This unique interplay allows us to explore various regimes of nonlinearity and phase transitions. Here, we observe a phase transition from the state in the regime of bistability to that in multistability in strongly interacting Rydberg atoms by varying the microwave field intensity, accompanying with the breaking of Z3-symmetry. During the phase transition, the system experiences a hidden critical point, in which the multistable states are difficult to be identified. Through changing the initial state of system, we can identify a hidden multistable state and reveal a hidden trajectory of phase transition, allowing us to track to a hidden critical point. In addition, we observe multiple phase transitions in spectra, suggesting higher-order symmetry breaking. The reported results shed light on manipulating multistability in dissipative Rydberg atoms systems and hold promise in the applications of non-equilibrium many-body physics.

cond-mat.quant-gas↗

arXiv2024

Tournaments with a Standard

We study tournaments where winning a rank-dependent prize requires passing a minimum performance standard. We show that, for any prize allocation, the optimal standard is always at a mode of performance that is weakly higher than the global mode and identify a necessary and sufficient condition for it to be at the global mode. When the prize scheme can be designed as well, the winner-take-all prize scheme is optimal for noise distributions with an increasing failure rate; and awarding equal prizes to all qualifying agents is optimal for noise distributions with a decreasing failure rate. For distributions with monotone likelihood ratios -- log-concave and log-convex, respectively -- these pay schemes are also optimal in a larger class of anonymous, monotone contracts that may depend on cardinal performance.

econ.TH↗

arXiv2024

Siamese Machine Unlearning with Knowledge Vaporization and Concentration

In response to the practical demands of the ``right to be forgotten" and the removal of undesired data, machine unlearning emerges as an essential technique to remove the learned knowledge of a fraction of data points from trained models. However, existing methods suffer from limitations such as insufficient methodological support, high computational complexity, and significant memory demands. In this work, we propose the concepts of knowledge vaporization and concentration to selectively erase learned knowledge from specific data points while maintaining representations for the remaining data. Utilizing the Siamese networks, we exemplify the proposed concepts and develop an efficient method for machine unlearning. Our proposed Siamese unlearning method does not require additional memory overhead and full access to the remaining dataset. Extensive experiments conducted across multiple unlearning scenarios showcase the superiority of Siamese unlearning over baseline methods, illustrating its ability to effectively remove knowledge from forgetting data, enhance model utility on remaining data, and reduce susceptibility to membership inference attacks.

cs.LG↗

arXiv2024

Toward Real-Time Edge AI: Model-Agnostic Task-Oriented Communication with Visual Feature Alignment

Task-oriented communication presents a promising approach to improve the communication efficiency of edge inference systems by optimizing learning-based modules to extract and transmit relevant task information. However, real-time applications face practical challenges, such as incomplete coverage and potential malfunctions of edge servers. This situation necessitates cross-model communication between different inference systems, enabling edge devices from one service provider to collaborate effectively with edge servers from another. Independent optimization of diverse edge systems often leads to incoherent feature spaces, which hinders the cross-model inference for existing task-oriented communication. To facilitate and achieve effective cross-model task-oriented communication, this study introduces a novel framework that utilizes shared anchor data across diverse systems. This approach addresses the challenge of feature alignment in both server-based and on-device scenarios. In particular, by leveraging the linear invariance of visual features, we propose efficient server-based feature alignment techniques to estimate linear transformations using encoded anchor data features. For on-device alignment, we exploit the angle-preserving nature of visual features and propose to encode relative representations with anchor data to streamline cross-model communication without additional alignment procedures during the inference. The experimental results on computer vision benchmarks demonstrate the superior performance of the proposed feature alignment approaches in cross-model task-oriented communications. The runtime and computation overhead analysis further confirm the effectiveness of the proposed feature alignment approaches in real-time applications.

cs.CV↗

arXiv2024

How to Collaborate: Towards Maximizing the Generalization Performance in Cross-Silo Federated Learning

Federated learning (FL) has attracted vivid attention as a privacy-preserving distributed learning framework. In this work, we focus on cross-silo FL, where clients become the model owners after training and are only concerned about the model's generalization performance on their local data. Due to the data heterogeneity issue, asking all the clients to join a single FL training process may result in model performance degradation. To investigate the effectiveness of collaboration, we first derive a generalization bound for each client when collaborating with others or when training independently. We show that the generalization performance of a client can be improved only by collaborating with other clients that have more training data and similar data distribution. Our analysis allows us to formulate a client utility maximization problem by partitioning clients into multiple collaborating groups. A hierarchical clustering-based collaborative training (HCCT) scheme is then proposed, which does not need to fix in advance the number of groups. We further analyze the convergence of HCCT for general non-convex loss functions which unveils the effect of data similarity among clients. Extensive simulations show that HCCT achieves better generalization performance than baseline schemes, whereas it degenerates to independent training and conventional FL in specific scenarios.

cs.LG↗

arXiv2024

A facile route to synthesize cubic gauche polymeric nitrogen

In this work, the long-sought cg-N with N-N single bond has been synthesized for the first time by a thermal-driven-only chemical route at ambient conditions. The successful synthesis of cg-N was achieved by first creating a solution of azides, which was then pretreated under vacuum conditions. Following the pretreatment, the resultant concentrated azide was heated at temperatures ranging from 260°C to 330°C for a reaction time of 3 hours, ultimately leading to the formation of cg-N. The emergent intense Raman peak characterized of cg-N provides solid evidence that the double bonded nitrogen-nitrogen transforms into a single bond form, which agrees well with cg-N structure. To date, this is the only work achieving the quantity of cg-N synthesized at ambient conditions by a facile route that can be further developed for the scalable synthesis and applications of polymerized nitrogen-based materials as high energy density materials.

cond-mat.mtrl-sci↗

arXiv2024

Federated Low-Rank Adaptation with Differential Privacy over Wireless Networks

Fine-tuning large pre-trained foundation models (FMs) on distributed edge devices presents considerable computational and privacy challenges. Federated fine-tuning (FedFT) mitigates some privacy issues by facilitating collaborative model training without the need to share raw data. To lessen the computational burden on resource-limited devices, combining low-rank adaptation (LoRA) with federated learning enables parameter-efficient fine-tuning. Additionally, the split FedFT architecture partitions an FM between edge devices and a central server, reducing the necessity for complete model deployment on individual devices. However, the risk of privacy eavesdropping attacks in FedFT remains a concern, particularly in sensitive areas such as healthcare and finance. In this paper, we propose a split FedFT framework with differential privacy (DP) over wireless networks, where the inherent wireless channel noise in the uplink transmission is utilized to achieve DP guarantees without adding an extra artificial noise. We shall investigate the impact of the wireless noise on convergence performance of the proposed framework. We will also show that by updating only one of the low-rank matrices in the split FedFT with DP, the proposed method can mitigate the noise amplification effect. Simulation results will demonstrate that the proposed framework achieves higher accuracy under strict privacy budgets compared to baseline methods.

cs.LG↗

arXiv2024

SALMONN-omni: A Codec-free LLM for Full-duplex Speech Understanding and Generation

Full-duplex multimodal large language models (LLMs) provide a unified framework for addressing diverse speech understanding and generation tasks, enabling more natural and seamless human-machine conversations. Unlike traditional modularised conversational AI systems, which separate speech recognition, understanding, and text-to-speech generation into distinct components, multimodal LLMs operate as single end-to-end models. This streamlined design eliminates error propagation across components and fully leverages the rich non-verbal information embedded in input speech signals. We introduce SALMONN-omni, a codec-free, full-duplex speech understanding and generation model capable of simultaneously listening to its own generated speech and background sounds while speaking. To support this capability, we propose a novel duplex spoken dialogue framework incorporating a ``thinking'' mechanism that facilitates asynchronous text and speech generation relying on embeddings instead of codecs (quantized speech and audio tokens). Experimental results demonstrate SALMONN-omni's versatility across a broad range of streaming speech tasks, including speech recognition, speech enhancement, and spoken question answering. Additionally, SALMONN-omni excels at managing turn-taking, barge-in, and echo cancellation scenarios, establishing its potential as a robust prototype for full-duplex conversational AI systems. To the best of our knowledge, SALMONN-omni is the first codec-free model of its kind. A full technical report along with model checkpoints will be released soon.

eess.AS↗

arXiv2024

Exploring Selective Layer Fine-Tuning in Federated Learning

Federated learning (FL) has emerged as a promising paradigm for fine-tuning foundation models using distributed data in a privacy-preserving manner. Under limited computational resources, clients often find it more practical to fine-tune a selected subset of layers, rather than the entire model, based on their task-specific data. In this study, we provide a thorough theoretical exploration of selective layer fine-tuning in FL, emphasizing a flexible approach that allows the clients to adjust their selected layers according to their local data and resources. We theoretically demonstrate that the layer selection strategy has a significant impact on model convergence in two critical aspects: the importance of selected layers and the heterogeneous choices across clients. Drawing from these insights, we further propose a strategic layer selection method that utilizes local gradients and regulates layer selections across clients. The extensive experiments on both image and text datasets demonstrate the effectiveness of the proposed strategy compared with several baselines, highlighting its advances in identifying critical layers that adapt to the client heterogeneity and training dynamics in FL.

cs.LG↗