Searcharxiv⌕ Search

arXiv subjects

Jian He

Publications and source records attributed to Jian He.

At least 37 records · Page 2Linked to original sources
arXiv2025

UB-Mesh: a Hierarchically Localized nD-FullMesh Datacenter Network Architecture

As the Large-scale Language Models (LLMs) continue to scale, the requisite computational power and bandwidth escalate. To address this, we introduce UB-Mesh, a novel AI datacenter network architecture designed to enhance scalability, performance, cost-efficiency and availability. Unlike traditional datacenters that provide symmetrical node-to-node bandwidth, UB-Mesh employs a hierarchically localized nD-FullMesh network topology. This design fully leverages the data locality of LLM training, prioritizing short-range, direct interconnects to minimize data movement distance and reduce switch usage. Although UB-Mesh's nD-FullMesh topology offers several theoretical advantages, its concrete architecture design, physical implementation and networking system optimization present new challenges. For the actual construction of UB-Mesh, we first design the UB-Mesh-Pod architecture, which is based on a 4D-FullMesh topology. UB-Mesh-Pod is implemented via a suite of hardware components that serve as the foundational building blocks, including specifically-designed NPU, CPU, Low-Radix-Switch (LRS), High-Radix-Switch (HRS), NICs and others. These components are interconnected via a novel Unified Bus (UB) technique, which enables flexible IO bandwidth allocation and hardware resource pooling. For networking system optimization, we propose advanced routing mechanism named All-Path-Routing (APR) to efficiently manage data traffic. These optimizations, combined with topology-aware performance enhancements and robust reliability measures like 64+1 backup design, result in 2.04x higher cost-efficiency, 7.2% higher network availability compared to traditional Clos architecture and 95%+ linearity in various LLM training tasks.

cs.AR↗
arXiv2025

Connecting Feedback to Choice: Understanding Educator Preferences in GenAI vs. Human-Created Lesson Plans in K-12 Education -- A Comparative Analysis

As generative AI (GenAI) models are increasingly explored for educational applications, understanding educator preferences for AI-generated lesson plans is critical for their effective integration into K-12 instruction. This exploratory study compares lesson plans authored by human curriculum designers, a fine-tuned LLaMA-2-13b model trained on K-12 content, and a customized GPT-4 model to evaluate their pedagogical quality across multiple instructional measures: warm-up activities, main tasks, cool-down activities, and overall quality. Using a large-scale preference study with K-12 math educators, we examine how preferences vary across grade levels and instructional components. We employ both qualitative and quantitative analyses. The raw preference results indicate that human-authored lesson plans are generally favored, particularly for elementary education, where educators emphasize student engagement, scaffolding, and collaborative learning. However, AI-generated models demonstrate increasing competitiveness in cool-down tasks and structured learning activities, particularly in high school settings. Beyond quantitative results, we conduct thematic analysis using LDA and manual coding to identify key factors influencing educator preferences. Educators value human-authored plans for their nuanced differentiation, real-world contextualization, and student discourse facilitation. Meanwhile, AI-generated lesson plans are often praised for their structure and adaptability for specific instructional tasks. Findings suggest a human-AI collaborative approach to lesson planning, where GenAI can serve as an assistive tool rather than a replacement for educator expertise in lesson planning. This study contributes to the growing discourse on responsible AI integration in education, highlighting both opportunities and challenges in leveraging GenAI for curriculum development.

cs.CY↗
arXiv2025

Diagnosis of Pulmonary Hypertension by Integrating Multimodal Data with a Hybrid Graph Convolutional and Transformer Network

Early and accurate diagnosis of pulmonary hypertension (PH) is essential for optimal patient management. Differentiating between pre-capillary and post-capillary PH is critical for guiding treatment decisions. This study develops and validates a deep learning-based diagnostic model for PH, designed to classify patients as non-PH, pre-capillary PH, or post-capillary PH. This retrospective study analyzed data from 204 patients (112 with pre-capillary PH, 32 with post-capillary PH, and 60 non-PH controls) at the First Affiliated Hospital of Nanjing Medical University. Diagnoses were confirmed through right heart catheterization. We selected 6 samples from each category for the test set (18 samples, 10%), with the remaining 186 samples used for the training set. This process was repeated 35 times for testing. This paper proposes a deep learning model that combines Graph convolutional networks (GCN), Convolutional neural networks (CNN), and Transformers. The model was developed to process multimodal data, including short-axis (SAX) sequences, four-chamber (4CH) sequences, and clinical parameters. Our model achieved a performance of Area under the receiver operating characteristic curve (AUC) = 0.81 +- 0.06(standard deviation) and Accuracy (ACC) = 0.73 +- 0.06 on the test set. The discriminative abilities were as follows: non-PH subjects (AUC = 0.74 +- 0.11), pre-capillary PH (AUC = 0.86 +- 0.06), and post-capillary PH (AUC = 0.83 +- 0.10). It has the potential to support clinical decision-making by effectively integrating multimodal data to assist physicians in making accurate and timely diagnoses.

eess.IV↗
arXiv2025

Adapting to Educate: Conversational AI's Role in Mathematics Education Across Different Educational Contexts

As educational settings increasingly integrate artificial intelligence (AI), understanding how AI tools identify -- and adapt their responses to -- varied educational contexts becomes paramount. This study examines conversational AI's effectiveness in supporting K-12 mathematics education across various educational contexts. Through qualitative content analysis, we identify educational contexts and key instructional needs present in educator prompts and assess AI's responsiveness. Our findings indicate that educators focus their AI conversations on assessment methods, how to set the cognitive demand level of their instruction, and strategies for making meaningful real-world connections. However, educators' conversations with AI about instructional practices do vary across revealed educational contexts; they shift their emphasis to tailored, rigorous content that addresses their students' unique needs. Educators often seek actionable guidance from AI and reject responses that do not align with their inquiries. While AI can provide accurate, relevant, and useful information when educational contexts or instructional practices are specified in conversation queries, its ability to consistently adapt responses along these evaluation dimensions varies across different educational settings. Significant work remains to realize the response-differentiating potential of conversational AI tools in complex educational use cases. This research contributes insights into developing AI tools that are responsive, proactive, and anticipatory, adapting to evolving educational needs before they are explicitly stated, and provides actionable recommendations for both developers and educators to enhance AI integration in educational practices.

cs.HC↗
arXiv2025

Milmer: a Framework for Multiple Instance Learning based Multimodal Emotion Recognition

Emotions play a crucial role in human behavior and decision-making, making emotion recognition a key area of interest in human-computer interaction (HCI). This study addresses the challenges of emotion recognition by integrating facial expression analysis with electroencephalogram (EEG) signals, introducing a novel multimodal framework-Milmer. The proposed framework employs a transformer-based fusion approach to effectively integrate visual and physiological modalities. It consists of an EEG preprocessing module, a facial feature extraction and balancing module, and a cross-modal fusion module. To enhance visual feature extraction, we fine-tune a pre-trained Swin Transformer on emotion-related datasets. Additionally, a cross-attention mechanism is introduced to balance token representation across modalities, ensuring effective feature integration. A key innovation of this work is the adoption of a multiple instance learning (MIL) approach, which extracts meaningful information from multiple facial expression images over time, capturing critical temporal dynamics often overlooked in previous studies. Extensive experiments conducted on the DEAP dataset demonstrate the superiority of the proposed framework, achieving a classification accuracy of 96.72% in the four-class emotion recognition task. Ablation studies further validate the contributions of each module, highlighting the significance of advanced feature extraction and fusion strategies in enhancing emotion recognition performance. Our code are available at https://github.com/liangyubuaa/Milmer.

cs.CV↗
arXiv2025

An Integrated Artificial Intelligence Operating System for Advanced Low-Altitude Aviation Applications

This paper introduces a high-performance artificial intelligence operating system tailored for low-altitude aviation, designed to address key challenges such as real-time task execution, computational efficiency, and seamless modular collaboration. Built on a powerful hardware platform and leveraging the UNIX architecture, the system implements a distributed data processing strategy that ensures rapid and efficient synchronization across critical modules, including vision, navigation, and perception. By adopting dynamic resource management, it optimally allocates computational resources, such as CPU and GPU, based on task priority and workload, ensuring high performance for demanding tasks like real-time video processing and AI model inference. Furthermore, the system features an advanced interrupt handling mechanism that allows for quick responses to sudden environmental changes, such as obstacle detection, by prioritizing critical tasks, thus improving safety and mission success rates. Robust security measures, including data encryption, access control, and fault tolerance, ensure the system's resilience against external threats and its ability to recover from potential hardware or software failures. Complementing these core features are modular components for image analysis, multi-sensor fusion, dynamic path planning, multi-drone coordination, and ground station monitoring. Additionally, a low-code development platform simplifies user customization, making the system adaptable to various mission-specific needs. This comprehensive approach ensures the system meets the evolving demands of intelligent aviation, providing a stable, efficient, and secure environment for complex drone operations.

cs.LG↗
arXiv2024

Efficient MedSAMs: Segment Anything in Medical Images on Laptop

Promptable segmentation foundation models have emerged as a transformative approach to addressing the diverse needs in medical images, but most existing models require expensive computing, posing a big barrier to their adoption in clinical practice. In this work, we organized the first international competition dedicated to promptable medical image segmentation, featuring a large-scale dataset spanning nine common imaging modalities from over 20 different institutions. The top teams developed lightweight segmentation foundation models and implemented an efficient inference pipeline that substantially reduced computational requirements while maintaining state-of-the-art segmentation accuracy. Moreover, the post-challenge phase advanced the algorithms through the design of performance booster and reproducibility tasks, resulting in improved algorithms and validated reproducibility of the winning solution. Furthermore, the best-performing algorithms have been incorporated into the open-source software with a user-friendly interface to facilitate clinical adoption. The data and code are publicly available to foster the further development of medical image segmentation foundation models and pave the way for impactful real-world applications.

eess.IV↗
arXiv2024

The tensorial description of the Auslander algebras of representation-finite string algebras

The aim of this article is to study the Auslander algebra of any representation-finite string algebra. More precisely, we introduce the notion of gluing algebras and show that the Auslander algebra of a representation-finite string algebra is a quotient of a \gluing algebra of $\vec{A}^e_n $. As applications, the Auslander algebras of two classes of string algebras whose quivers are Dynkin types $A$ and $D$ are described. Moreover, the representation types of the above Auslander algebras are also given exactly.

math.RT↗
arXiv2024

Highly Antioxidative Lithium Salt Enables High-Voltage Ether Electrolyte for Lithium Metal Battery

Ether-based electrolytes exhibit excellent compatibility with Li metal anodes, but their instability at high voltages limits their use in high-voltage Li metal batteries. To address this issue, we introduce an alternative perfluorobutane sulfonate (LiPFBS) / dimethoxyethane (DME) electrolyte to stabilize DME in a 4.6 V Li-Li LCO battery. Our study focuses on the formation of solid-electrolyte interphase (SEI) and cathode-electrolyte interphase (CEI) layers compared to the LiTFSI/DME electrolyte. We demonstrate that LiPFBS helps maintain DME's compatibility in SEI formation. Additionally, a durable CEI layer derived from PFBS$^{-}$ enhances the performance of the cell at high voltages by forming a robust, inorganic-dominant CEI layer. A PFBS$^{-}$-derived CEI significantly enhances the overall performance of the full cell under high voltage conditions.

cond-mat.mtrl-sci↗
arXiv2024

Higher Auslander-Reiten sequences revisited

Let $(\mathscr{C},\mathbb{E},\mathfrak{s})$ be an $n$-exangulated category with enough projectives and enough injectives, and $\mathscr{X}$ be a cluster-tilting subcategory of $\mathscr{C}$. Liu and Zhou have shown that the quotient category $\mathscr{C}/\mathscr{X}$ is an $n$-abelian category. In this paper, we prove that if $\mathscr{C}$ has Auslander-Reiten $n$-exangles, then $\mathscr{C}/\mathscr{X}$ has Auslander-Reiten $n$-exact sequences. Moreover, we also show that if a Frobenius $n$-exangulated category $\mathscr{C}$ has Auslander-Reiten $n$-exangles, then the stable category $\overline{\mathscr{C}}$ of $\mathscr{C}$ has Auslander-Reiten $(n+2)$-angles.

math.RT↗
arXiv2024

Injectivity of modules over trusses

As the dual notion of projective modules over trusses, injective modules over trusses are introduced. The Schanuel Lemmas on projective and injective modules over trusses are exhibited in this paper.

math.RT↗
arXiv2024

The structure of Hurwitz numbers with fixed ramification profile and varying genus

In 1891, Hurwitz introduced the enumeration of genus $g$, degree $d$, branched covers of the Riemann sphere with simple ramification over prescribed points and no branching elsewhere. He showed that for fixed degree $d$, the enumeration possesses a remarkable structure. More precisely, it can be expressed as a linear combination of exponentials $m^{2g-2+2d}$, where $m$ ranges over the integers from $1$ to $\binom{d}{2}$. In this paper, we generalise this structural result to Hurwitz numbers that enumerate branched covers which also have a prescribed ramification profile over one point. Our proof fundamentally uses the infinite wedge space, in particular the connected correlators of products of $\mathcal{E}$-operators. The recent study of Hurwitz numbers has often focussed on their structure with fixed genus and varying ramification profile. Our main result is orthogonal to this, allowing for the explicit calculation and the asymptotic analysis of Hurwitz numbers in large genus. We pose the broad question of which other enumerative problems exhibit analogous structure. We prove that orbifold Hurwitz numbers can also be expressed as a linear combination of exponentials and conjecture that monotone Hurwitz numbers share a similar structure, but with the inclusion of an additional linear term.

math.CO↗
arXiv2024

Automatic Organ and Pan-cancer Segmentation in Abdomen CT: the FLARE 2023 Challenge

Organ and cancer segmentation in abdomen Computed Tomography (CT) scans is the prerequisite for precise cancer diagnosis and treatment. Most existing benchmarks and algorithms are tailored to specific cancer types, limiting their ability to provide comprehensive cancer analysis. This work presents the first international competition on abdominal organ and pan-cancer segmentation by providing a large-scale and diverse dataset, including 4650 CT scans with various cancer types from over 40 medical centers. The winning team established a new state-of-the-art with a deep learning-based cascaded framework, achieving average Dice Similarity Coefficient scores of 92.3% for organs and 64.9% for lesions on the hidden multi-national testing set. The dataset and code of top teams are publicly available, offering a benchmark platform to drive further innovations https://codalab.lisn.upsaclay.fr/competitions/12239.

eess.IV↗
arXiv2024

YZS-model: A Predictive Model for Organic Drug Solubility Based on Graph Convolutional Networks and Transformer-Attention

Accurate prediction of drug molecule solubility is crucial for therapeutic effectiveness and safety. Traditional methods often miss complex molecular structures, leading to inaccuracies. We introduce the YZS-Model, a deep learning framework integrating Graph Convolutional Networks (GCN), Transformer architectures, and Long Short-Term Memory (LSTM) networks to enhance prediction precision. GCNs excel at capturing intricate molecular topologies by modeling the relationships between atoms and bonds. Transformers, with their self-attention mechanisms, effectively identify long-range dependencies within molecules, capturing global interactions. LSTMs process sequential data, preserving long-term dependencies and integrating temporal information within molecular sequences. This multifaceted approach leverages the strengths of each component, resulting in a model that comprehensively understands and predicts molecular properties. Trained on 9,943 compounds and tested on an anticancer dataset, the YZS-Model achieved an $R^2$ of 0.59 and an RMSE of 0.57, outperforming benchmark models ($R^2$ of 0.52 and RMSE of 0.61). In an independent test, it demonstrated an RMSE of 1.05, improving accuracy by 45.9%. The integration of these deep learning techniques allows the YZS-Model to learn valuable features from complex data without predefined parameters, handle large datasets efficiently, and adapt to various molecular types. This comprehensive capability significantly improves predictive accuracy and model generalizability. Its precision in solubility predictions can expedite drug development by optimizing candidate selection, reducing costs, and enhancing efficiency. Our research underscores deep learning's transformative potential in pharmaceutical science, particularly for solubility prediction and drug design.

cs.LG↗
arXiv2024

Calibration of the rating transition model for high and low default portfolios

In this paper we develop Maximum likelihood (ML) based algorithms to calibrate the model parameters in credit rating transition models. Since the credit rating transition models are not Gaussian linear models, the celebrated Kalman filter is not suitable to compute the likelihood of observed migrations. Therefore, we develop a Laplace approximation of the likelihood function and as a result the Kalman filter can be used in the end to compute the likelihood function. This approach is applied to so-called high-default portfolios, in which the number of migrations (defaults) is large enough to obtain high accuracy of the Laplace approximation. By contrast, low-default portfolios have a limited number of observed migrations (defaults). Therefore, in order to calibrate low-default portfolios, we develop a ML algorithm using a particle filter (PF) and Gaussian process regression. Experiments show that both algorithms are efficient and produce accurate approximations of the likelihood function and the ML estimates of the model parameters.

q-fin.RM↗
arXiv2024

$n$-cotorsion pairs and recollements of extriangulated categories

In this article, we prove that if $(\mathcal A ,\mathcal B,\mathcal C)$ is a recollement of extriangulated categories, then $n$-cotorsion pairs in $\mathcal A$ and $\mathcal C$ can induce $n$-cotorsion pairs in $\mathcal B$. Conversely, this holds true under natural assumptions. Besides, we give mild conditions on a pseudo cluster tilting subcategory on the middle category of a recollement of extriangulated categories, for the corresponding additive quotients to form a recollement of semi-abelian categories.

math.RT↗
arXiv2024

Slip topology of steady flows around a critical point: Taking the linear velocity field as an example

The flow of viscous fluids is considered as the aggregation of the motion of fluid particles when the fluid is conceived to be made up by an infinite number of particles. As an alternative of this conventional model, fluid motion could be understood as the slip of fluid layers with a molecular scale over each other, where the slip structures of fluid and their associated small-scale motion are characterized by an axial-vector-valued differential 1-form, called the vortex field. In this paper, in the case of steady flows we define the swirling degree of the velocity field at a point, and further the swirl field of the steady flow, to study the slip topology of fluid or the local streamline pattern around the critical point. The linear velocity field in the right real Schur form is used to carry out detailed analyses around the isolated critical point. Theoretical deduction and numerical test unveil the connection between the swirling degree and the swirl field, greatly make clear the topological property of slip structures of fluid in steady flows, especially in three-dimensional space.

physics.flu-dyn↗