arXiv subjects

Wei Jiang

Publications and source records attributed to Wei Jiang.

At least 145 records · Page 8Linked to original sources

arXiv2025

Convergence Analysis of the Lion Optimizer in Centralized and Distributed Settings

In this paper, we analyze the convergence properties of the Lion optimizer. First, we establish that the Lion optimizer attains a convergence rate of $\mathcal{O}(d^{1/2}T^{-1/4})$ under standard assumptions, where $d$ denotes the problem dimension and $T$ is the iteration number. To further improve this rate, we introduce the Lion optimizer with variance reduction, resulting in an enhanced convergence rate of $\mathcal{O}(d^{1/2}T^{-1/3})$. We then analyze in distributed settings, where the standard and variance reduced version of the distributed Lion can obtain the convergence rates of $\mathcal{O}(d^{1/2}(nT)^{-1/4})$ and $\mathcal{O}(d^{1/2}(nT)^{-1/3})$, with $n$ denoting the number of nodes. Furthermore, we investigate a communication-efficient variant of the distributed Lion that ensures sign compression in both communication directions. By employing the unbiased sign operations, the proposed Lion variant and its variance reduction counterpart, achieve convergence rates of $\mathcal{O}\left( \max \left\{\frac{d^{1/4}}{T^{1/4}}, \frac{d^{1/10}}{n^{1/5}T^{1/5}} \right\} \right)$ and $\mathcal{O}\left( \frac{d^{1/4}}{T^{1/4}} \right)$, respectively.

cs.LG↗

arXiv2025

ASM-UNet: Adaptive Scan Mamba Integrating Group Commonalities and Individual Variations for Fine-Grained Segmentation

Precise lesion resection depends on accurately identifying fine-grained anatomical structures. While many coarse-grained segmentation (CGS) methods have been successful in large-scale segmentation (e.g., organs), they fall short in clinical scenarios requiring fine-grained segmentation (FGS), which remains challenging due to frequent individual variations in small-scale anatomical structures. Although recent Mamba-based models have advanced medical image segmentation, they often rely on fixed manually-defined scanning orders, which limit their adaptability to individual variations in FGS. To address this, we propose ASM-UNet, a novel Mamba-based architecture for FGS. It introduces adaptive scan scores to dynamically guide the scanning order, generated by combining group-level commonalities and individual-level variations. Experiments on two public datasets (ACDC and Synapse) and a newly proposed challenging biliary tract FGS dataset, namely BTMS, demonstrate that ASM-UNet achieves superior performance in both CGS and FGS tasks. Our code and dataset are available at https://github.com/YqunYang/ASM-UNet.

cs.CV↗

arXiv2025

An Evolving Scenario Generation Method based on Dual-modal Driver Model Trained by Multi-Agent Reinforcement Learning

In the autonomous driving testing methods based on evolving scenarios, the construction method of the driver model, which determines the driving maneuvers of background vehicles (BVs) in the scenario, plays a critical role in generating safety-critical scenarios. In particular, the cooperative adversarial driving characteristics between BVs can contribute to the efficient generation of safety-critical scenarios with high testing value. In this paper, a multi-agent reinforcement learning (MARL) method is used to train and generate a dual-modal driver model (Dual-DM) with non-adversarial and adversarial driving modalities. The model is then connected to a continuous simulated traffic environment to generate complex, diverse and strong interactive safety-critical scenarios through evolving scenario generation method. After that, the generated evolving scenarios are evaluated in terms of fidelity, test efficiency, complexity and diversity. Results show that without performance degradation in scenario fidelity (>85% similarity to real-world scenarios) and complexity (complexity metric: 0.45, +32.35% and +12.5% over two baselines), Dual-DM achieves a substantial enhancement in the efficiency of generating safety-critical scenarios (efficiency metric: 0.86, +195% over two baselines). Furthermore, statistical analysis and case studies demonstrate the diversity of safety-critical evolving scenarios generated by Dual-DM in terms of the adversarial interaction patterns. Therefore, Dual-DM can greatly improve the performance of the generation of safety-critical scenarios through evolving scenario generation method.

cs.LG↗

arXiv2025

Dual Adaptivity: Universal Algorithms for Minimizing the Adaptive Regret of Convex Functions

To deal with changing environments, a new performance measure -- adaptive regret, defined as the maximum static regret over any interval, was proposed in online learning. Under the setting of online convex optimization, several algorithms have been successfully developed to minimize the adaptive regret. However, existing algorithms lack universality in the sense that they can only handle one type of convex functions and need apriori knowledge of parameters, which hinders their application in real-world scenarios. To address this limitation, this paper investigates universal algorithms with dual adaptivity, which automatically adapt to the property of functions (convex, exponentially concave, or strongly convex), as well as the nature of environments (stationary or changing). Specifically, we propose a meta-expert framework for dual adaptive algorithms, where multiple experts are created dynamically and aggregated by a meta-algorithm. The meta-algorithm is required to yield a second-order bound, which can accommodate unknown function types. We further incorporate the technique of sleeping experts to capture the changing environments. For the construction of experts, we introduce two strategies (increasing the number of experts or enhancing the capabilities of experts) to achieve universality. Theoretical analysis shows that our algorithms are able to minimize the adaptive regret for multiple types of convex functions simultaneously, and also allow the type of functions to switch between rounds. Moreover, we extend our meta-expert framework to online composite optimization, and develop a universal algorithm for minimizing the adaptive regret of composite functions.

cs.LG↗

arXiv2025

Quantitative Benchmarking of Remote Excitation in Plasmonic Sensing with Enhanced Signal-to-Noise Ratio

Remote excitation using guided optical modes -- such as waveguides, fibers, or surface waves -- offers a promising alternative to direct optical excitation for surface-enhanced Raman scattering (SERS), particularly in applications requiring reduced heating, minimal invasiveness, and on-chip integration. However, despite its widespread use, systematic comparisons between remote and direct excitation remain limited. Here, we quantitatively benchmark both schemes by measuring power-dependent SERS responses from individual plasmonic nanogaps. We statistically analyze the maximum achievable SERS intensity before structural degradation, extract local temperatures, and evaluate signal-to-noise ratios (SNR). Our findings reveal that both remote and direct SERS share a common electric-field limit, despite exhibiting different levels of heating. This suggests that spectral evolution is primarily governed by the local electric field, which drives nanoscale atomic migration rather than excessive heating. Nonetheless, the lower heating associated with remote excitation enhances the Raman SNR by approximately 30%, improving measurement quality without compromising signal strength. This study establishes a quantitative framework for evaluating excitation strategies in plasmonic sensing, and challenges common assumptions about the role of heating in nanostructural stability under strong optical excitation.

physics.optics↗

arXiv2025

Epidemiology-informed Network for Robust Rumor Detection

The rapid spread of rumors on social media has posed significant challenges to maintaining public trust and information integrity. Since an information cascade process is essentially a propagation tree, recent rumor detection models leverage graph neural networks to additionally capture information propagation patterns, thus outperforming text-only solutions. Given the variations in topics and social impact of the root node, different source information naturally has distinct outreach capabilities, resulting in different heights of propagation trees. This variation, however, impedes the data-driven design of existing graph-based rumor detectors. Given a shallow propagation tree with limited interactions, it is unlikely for graph-based approaches to capture sufficient cascading patterns, questioning their ability to handle less popular news or early detection needs. In contrast, a deep propagation tree is prone to noisy user responses, and this can in turn obfuscate the predictions. In this paper, we propose a novel Epidemiology-informed Network (EIN) that integrates epidemiological knowledge to enhance performance by overcoming data-driven methods sensitivity to data quality. Meanwhile, to adapt epidemiology theory to rumor detection, it is expected that each users stance toward the source information will be annotated. To bypass the costly and time-consuming human labeling process, we take advantage of large language models to generate stance labels, facilitating optimization objectives for learning epidemiology-informed representations. Our experimental results demonstrate that the proposed EIN not only outperforms state-of-the-art methods on real-world datasets but also exhibits enhanced robustness across varying tree depths.

cs.SI↗

arXiv2025

A new Phase-Field Model for Anisotropic Surface Diffusion: Anisotropic Cahn-Hilliard Equation with Improved Conservation (ACH-IC)

As popular approximations to sharp-interface models, the Cahn-Hilliard type phase-field models are usually used to simulate interface dynamics with volume conservation. However, the convergence rate of the volume enclosed by the interface to its sharp-interface limit is usually at first order of the interface thickness in the classical Cahn-Hilliard model with constant or degenerate mobilities. In this work, we propose a variational framework for developing new Cahn-Hilliard dynamics with enhanced volume conservation by introducing a more general conserved quantity. In particular, based on Onsager's variational principle (OVP) and a modified conservation law, we develop an anisotropic Cahn-Hilliard equation with improved conservation (ACH-IC) for approximating anisotropic surface diffusion. The ACH-IC model employs a new conserved quantity that approximates a step function more effectively, and yields second-order volume conservation while preserving energy dissipation for the classical anisotropic surface energy. The second-order volume conservation as well as the convergence to the sharp-interface surface diffusion dynamics is derived through comprehensive asymptotic analysis. Numerical evidence not only reveals the underlying physics of the proposed model in comparison with the classical one, but also demonstrates its exceptional performance in simulating anisotropic surface diffusion dynamics.

math-ph↗

arXiv2025

ECVC: Exploiting Non-Local Correlations in Multiple Frames for Contextual Video Compression

In Learned Video Compression (LVC), improving inter prediction, such as enhancing temporal context mining and mitigating accumulated errors, is crucial for boosting rate-distortion performance. Existing LVCs mainly focus on mining the temporal movements while neglecting non-local correlations among frames. Additionally, current contextual video compression models use a single reference frame, which is insufficient for handling complex movements. To address these issues, we propose leveraging non-local correlations across multiple frames to enhance temporal priors, significantly boosting rate-distortion performance. To mitigate error accumulation, we introduce a partial cascaded fine-tuning strategy that supports fine-tuning on full-length sequences with constrained computational resources. This method reduces the train-test mismatch in sequence lengths and significantly decreases accumulated errors. Based on the proposed techniques, we present a video compression scheme ECVC. Experiments demonstrate that our ECVC achieves state-of-the-art performance, reducing 10.5% and 11.5% more bit-rates than previous SOTA method DCVC-FM over VTM-13.2 low delay B (LDB) under the intra period (IP) of 32 and -1, respectively.

eess.IV↗

arXiv2025

BiECVC: Gated Diversification of Bidirectional Contexts for Learned Video Compression

Recent forward prediction-based learned video compression (LVC) methods have achieved impressive results, even surpassing VVC reference software VTM under the Low Delay B (LDB) configuration. In contrast, learned bidirectional video compression (BVC) remains underexplored and still lags behind its forward-only counterparts. This performance gap is mainly due to the limited ability to extract diverse and accurate contexts: most existing BVCs primarily exploit temporal motion while neglecting non-local correlations across frames. Moreover, they lack the adaptability to dynamically suppress harmful contexts arising from fast motion or occlusion. To tackle these challenges, we propose BiECVC, a BVC framework that incorporates diversified local and non-local context modeling along with adaptive context gating. For local context enhancement, BiECVC reuses high-quality features from lower layers and aligns them using decoded motion vectors without introducing extra motion overhead. To model non-local dependencies efficiently, we adopt a linear attention mechanism that balances performance and complexity. To further mitigate the impact of inaccurate context prediction, we introduce Bidirectional Context Gating, inspired by data-dependent decay in recent autoregressive language models, to dynamically filter contextual information based on conditional coding results. Extensive experiments demonstrate that BiECVC achieves state-of-the-art performance, reducing the bit-rate by 13.4% and 15.7% compared to VTM 13.2 under the Random Access (RA) configuration with intra periods of 32 and 64, respectively. To our knowledge, BiECVC is the first learned video codec to surpass VTM 13.2 RA across all standard test datasets.

eess.IV↗

arXiv2025

7B Fully Open Source Moxin-LLM/VLM -- From Pretraining to GRPO-based Reinforcement Learning Enhancement

Recently, Large Language Models (LLMs) have undergone a significant transformation, marked by a rapid rise in both their popularity and capabilities. Leading this evolution are proprietary LLMs like GPT-4 and GPT-o1, which have captured widespread attention in the AI community due to their remarkable performance and versatility. Simultaneously, open-source LLMs, such as LLaMA, have made great contributions to the ever-increasing popularity of LLMs due to the ease to customize and deploy the models across diverse applications. Although open-source LLMs present unprecedented opportunities for innovation and research, the commercialization of LLMs has raised concerns about transparency, reproducibility, and safety. Many open-source LLMs fail to meet fundamental transparency requirements by withholding essential components like training code and data, which may hinder further innovations on LLMs. To mitigate this issue, we introduce Moxin 7B, a fully open-source LLM developed, adhering to principles of open science, open source, open data, and open access. We release the pre-training code and configurations, training and fine-tuning datasets, and intermediate and final checkpoints, aiming to make continuous commitments to fully open-source LLMs. After pre-training the base model, we finetune the Moxin Base model with SOTA post-training framework and instruction data to obtain Moxin Instruct model. To improve the reasoning capability, we further finetune our Instruct model with chain-of-thought data distilled from DeepSeek R1, and then use Group Relative Policy Optimization (GRPO) following DeepSeek R1 to finetune our model, leading to the Moxin Reasoning model. Moreover, we develop our vision language model based on our Moxin model. Experiments show that our models achieve superior performance in various evaluations such as zero-shot evaluation, few-shot evaluation, and CoT evaluation.

cs.CL↗

arXiv2025

Efficiently Serving Large Multimodal Models Using EPD Disaggregation

Large Multimodal Models (LMMs) extend Large Language Models (LLMs) by handling diverse inputs such as images, audio, and video, but at the cost of adding a multimodal encoding stage that increases both computational and memory overhead. This step negatively affects key Service Level Objectives (SLOs), such as time to first token (TTFT) and time per output token (TPOT). We introduce Encode-Prefill-Decode (EPD) Disaggregation, a novel framework that separates the encoding, prefill, and decode stages onto dedicated resources. Unlike current systems, which bundle encoding and prefill together, our approach decouples these steps, unlocking new opportunities and optimizations. These include a mechanism to cache multimedia tokens for efficient transfer, a novel way to parallelize the encoding load within a request, a module for optimal resource allocation for disaggregated serving, and a novel role-switching method to handle changing workload characteristics. Experimental evaluations with popular LMMs show substantial gains in memory efficiency (up to 15x lower peak memory utilization), batch sizes (up to 22x larger), 10x more images per request, and 2.2x larger KV caches. Furthermore, it leads to significant improvements in SLO attainment (up to 90-100% improvement) and TTFT (up to 71% reduction), compared to systems that do not disaggregate. The code is available at https://github.com/vbdi/epdserve.

cs.DC↗

arXiv2025

Heterogeneous Massive MIMO: A Cost-Efficient Technique for Uniform Service in Cellular Networks

Massive multi-input multi-output (MIMO) has evolved along two tracks: cellular and cell-free, each with unique advantages and limitations. The cellular approach suffers from worse user spectral efficiency at cell edges, whereas the cell-free approach incurs high implementation costs due to a large-scale distributed infrastructure. This paper introduces a novel networking paradigm, termed heterogeneous massive MIMO (HmMIMO), which seamlessly integrates co-located and distributed antennas. Differing from two conventional paradigms, HmMIMO remains a base station with a large antenna array at the center of each cell, aided by distributed antennas deployed at cell edges. Our findings demonstrate that this paradigm achieves a favorable trade-off between performance and implementation complexity.

cs.IT↗

arXiv2025

Near-Field Propagation and Spatial Non-Stationarity Channel Model for 6-24 GHz (FR3) Extremely Large-Scale MIMO: Adopted by 3GPP for 6G

Next generation cellular deployments are expected to exploit the 6-24 GHz frequency range 3 (FR3) and extremely large-scale multiple-input multiple-output (XL-MIMO) to enable ultra-high data rates and reliability. However, the significantly enlarged antenna apertures and higher carrier frequencies render the far-field and spatial stationarity assumptions in the existing 3rd generation partnership project (3GPP) channel models invalid, giving rise to new features such as near-field propagation and spatial non-stationarity (SNS). Despite extensive prior research, incorporating these new features within the standardized channel modeling framework remains an open issue. To address this, this paper presents a channel modeling framework for XL-MIMO systems that incorporates both near-field and SNS features, adopted by 3GPP. For the near-field propagation feature, the framework models the distances from the base station (BS) and user equipment to the spherical-wave sources associated with clusters. These distances are used to characterize element-wise variations of path parameters, such as nonlinear changes in phase and angle. To capture the effect of SNS at the BS side, a stochastic-based approach is proposed to model SNS caused by incomplete scattering, by establishing power attenuation factors from visibility probability and visibility region to characterize antenna element-wise path power variation. In addition, a physical blocker-based approach is introduced to model SNS effects caused by partial blockage. Finally, a simulation framework for near-field and SNS is developed within the structure of the existing 3GPP channel model. Performance evaluations demonstrate that the near-field model captures higher channel capacity potential compared to the far-field model. Coupling loss results indicate that SNS leads to more pronounced propagation fading relative to the spatial stationary model.

eess.SP↗

arXiv2025

Giant Magneto-Optical Effects in Two-Dimensional Flat-Band Antiferromagnets

In this work, we reveal giant magneto-optical responses in two-dimensional(2D) antiferromagnets with nearly flat electronic bands, based on first-principles calculations and group-theoretical analysis. We identify a record-large second-order magneto-optical Schafer-Hubert(SH) effect, featuring a polarization rotation angle of 28 degree, in monolayer antiferromagnetic RuOCl2, driven by flatband-enhanced interband optical transitions. Both the valence and conduction bands exhibit pronounced directional flatness, giving rise to highly anisotropic optical absorption and broadband hyperbolic frequency windows spanning the entire visible spectrum. This anisotropy leads to an exceptionally strong linear dichroism (LD) reaching 50%, far exceeding values reported in other 2D magnetic systems. Remarkably, the giant SH effect and LD appear at distinct photon energies, reflecting a momentum-direction-dependent crossover between flat and dispersive bands. Both responses are further amplified with increasing RuOCl2 film thickness. Our results establish flat-band antiferromagnets as a fertile platform for realizing giant nonlinear magneto-optical effects and open new avenues for 2D opto-spintronic device applications.

cond-mat.mtrl-sci↗

arXiv2025

Aggregated Structural Representation with Large Language Models for Human-Centric Layout Generation

Time consumption and the complexity of manual layout design make automated layout generation a critical task, especially for multiple applications across different mobile devices. Existing graph-based layout generation approaches suffer from limited generative capability, often resulting in unreasonable and incompatible outputs. Meanwhile, vision based generative models tend to overlook the original structural information, leading to component intersections and overlaps. To address these challenges, we propose an Aggregation Structural Representation (ASR) module that integrates graph networks with large language models (LLMs) to preserve structural information while enhancing generative capability. This novel pipeline utilizes graph features as hierarchical prior knowledge, replacing the traditional Vision Transformer (ViT) module in multimodal large language models (MLLM) to predict full layout information for the first time. Moreover, the intermediate graph matrix used as input for the LLM is human editable, enabling progressive, human centric design generation. A comprehensive evaluation on the RICO dataset demonstrates the strong performance of ASR, both quantitatively using mean Intersection over Union (mIoU), and qualitatively through a crowdsourced user study. Additionally, sampling on relational features ensures diverse layout generation, further enhancing the adaptability and creativity of the proposed approach.

cs.CV↗

arXiv2025

CONCORD: Concept-Informed Diffusion for Dataset Distillation

Dataset distillation (DD) has witnessed significant progress in creating small datasets that encapsulate rich information from large original ones. Particularly, methods based on generative priors show promising performance, while maintaining computational efficiency and cross-architecture generalization. However, the generation process lacks explicit controllability for each sample. Previous distillation methods primarily match the real distribution from the perspective of the entire dataset, whereas overlooking concept completeness at the instance level. The missing or incorrectly represented object details cannot be efficiently compensated due to the constrained sample amount typical in DD settings. To this end, we propose incorporating the concept understanding of large language models (LLMs) to perform Concept-Informed Diffusion (CONCORD) for dataset distillation. Specifically, distinguishable and fine-grained concepts are retrieved based on category labels to inform the denoising process and refine essential object details. By integrating these concepts, the proposed method significantly enhances both the controllability and interpretability of the distilled image generation, without relying on pre-trained classifiers. We demonstrate the efficacy of CONCORD by achieving state-of-the-art performance on ImageNet-1K and its subsets. The code implementation is released in https://github.com/vimar-gu/CONCORD.

cs.CV↗

arXiv2025

Improved Methods for Model Pruning and Knowledge Distillation

Model pruning is a performance optimization technique for large language models like R1 or o3-mini. However, existing pruning methods often lead to significant performance degradation or require extensive retraining and fine-tuning. This technique aims to identify and remove neurons, connections unlikely leading to the contribution during the human-computer interaction phase. Our goal is to obtain a much smaller and faster knowledge distilled model that can quickly generate content almost as good as those of the unpruned ones. We propose MAMA Pruning, short for Movement and Magnitude Analysis, an improved pruning method that effectively reduces model size and computational complexity while maintaining performance comparable to the original unpruned model even at extreme pruned levels. The improved method is based on weights, bias fixed in the pre-training phase and GRPO rewards verified during the post-training phase as our novel pruning indicators. Preliminary experimental results show that our method outperforms and be comparable to state-of-the-art methods across various pruning levels and different downstream computational linguistics tasks.

cs.CL↗

arXiv2025

Multi-View Learning with Context-Guided Receptance for Image Denoising

Image denoising is essential in low-level vision applications such as photography and automated driving. Existing methods struggle with distinguishing complex noise patterns in real-world scenes and consume significant computational resources due to reliance on Transformer-based models. In this work, the Context-guided Receptance Weighted Key-Value (\M) model is proposed, combining enhanced multi-view feature integration with efficient sequence modeling. Our approach introduces the Context-guided Token Shift (CTS) paradigm, which effectively captures local spatial dependencies and enhance the model's ability to model real-world noise distributions. Additionally, the Frequency Mix (FMix) module extracting frequency-domain features is designed to isolate noise in high-frequency spectra, and is integrated with spatial representations through a multi-view learning process. To improve computational efficiency, the Bidirectional WKV (BiWKV) mechanism is adopted, enabling full pixel-sequence interaction with linear complexity while overcoming the causal selection constraints. The model is validated on multiple real-world image denoising datasets, outperforming the existing state-of-the-art methods quantitatively and reducing inference time up to 40\%. Qualitative results further demonstrate the ability of our model to restore fine details in various scenes.

eess.IV↗