arXiv subjects

Yan Li

Publications and source records attributed to Yan Li.

At least 217 records · Page 12Linked to original sources

arXiv2025

A reference frame-based microgrid primary control for ensuring global convergence to a periodic orbit

Power systems with a high penetration of renewable generation are vulnerable to frequency oscillation and voltage instability. Traditionally, the stability of power systems is considered either in terms of local stability or as an angle oscillator synchronization problem with the simplifying assumption that the dynamics of the amplitudes are on much shorter time scales. Without this assumption, however, the steady state being studied is essentially a limit cycle with the convergence of its orbit in question. In this paper, we present a method to analyze the orbital stability of a microgrid and propose a voltage controller for the inverter-interfaced renewable generators. The main hurdle to the problem lies in the constant terms in the rotating internal reference frames of each generator. We extend the shifted passivity of port-Hamiltonian systems to the analysis of limit cycles and prove that, if the system is shifted passive without considering these constant terms, then the periodic orbit is globally attractive. To the best of our knowledge, this is the first global stability result for non-nominal steady states of the microgrid in the full state space, which provides new insights into the synchronization phenomenon where the dissipativity of the system ensures convergence. The proposed controller is verified with a test microgrid, demonstrating its stability and transient smoothness compared to the standard droop control.

eess.SY↗

arXiv2025

Who Owns This Sample: Cross-Client Membership Inference Attack in Federated Graph Neural Networks

Graph-structured data is prevalent in many real-world applications, including social networks, financial systems, and molecular biology. Graph Neural Networks (GNNs) have become the de facto standard for learning from such data due to their strong representation capabilities. As GNNs are increasingly deployed in federated learning (FL) settings to preserve data locality and privacy, new privacy threats arise from the interaction between graph structures and decentralized training. In this paper, we present the first systematic study of cross-client membership inference attacks (CC-MIA) against node classification tasks of federated GNNs (FedGNNs), where a malicious client aims to infer which client owns the given data. Unlike prior centralized-focused work that focuses on whether a sample was included in training, our attack targets sample-to-client attribution, a finer-grained privacy risk unique to federated settings. We design a general attack framework that exploits FedGNNs' aggregation behaviors, gradient updates, and embedding proximity to link samples to their source clients across training rounds. We evaluate our attack across multiple graph datasets under realistic FL setups. Results show that our method achieves high performance on both membership inference and ownership identification. Our findings highlight a new privacy threat in federated graph learning-client identity leakage through structural and model-level cues, motivating the need for attribution-robust GNN design.

cs.LG↗

arXiv2025

Scheduzz: Constraint-based Fuzz Driver Generation with Dual Scheduling

Fuzzing a library requires experts to understand the library usage well and craft high-quality fuzz drivers, which is tricky and tedious. Therefore, many techniques have been proposed to automatically generate fuzz drivers. However, they fail to generate rational fuzz drivers due to the lack of adherence to proper library usage conventions, such as ensuring a resource is closed after being opened. To make things worse, existing library fuzzing techniques unconditionally execute each driver, resulting in numerous irrational drivers that waste computational resources while contributing little coverage and generating false positive bug reports. To tackle these challenges, we propose a novel automatic library fuzzing technique, Scheduzz, an LLM-based library fuzzing technique. It leverages LLMs to understand rational usage of libraries and extract API combination constraints. To optimize computational resource utilization, a dual scheduling framework is implemented to efficiently manage API combinations and fuzz drivers. The framework models driver generation and the corresponding fuzzing campaign as an online optimization problem. Within the scheduling loop, multiple API combinations are selected to generate fuzz drivers, while simultaneously, various optimized fuzz drivers are scheduled for execution or suspension. We implemented Scheduzz and evaluated it in 33 real-world libraries. Compared to baseline approaches, Scheduzz significantly reduces computational overhead and outperforms UTopia on 16 out of 21 libraries. It achieves 1.62x, 1.50x, and 1.89x higher overall coverage than the state-of-the-art techniques CKGFuzzer, Promptfuzz, and the handcrafted project OSS-Fuzz, respectively. In addition, Scheduzz discovered 33 previously unknown bugs in these well-tested libraries, 3 of which have been assigned CVEs.

cs.SE↗

arXiv2025

Analysis of Fe and Co binary catalysts in chemical vapor deposition growth of single-walled carbon nanotubes

Metal catalysts play a pivotal role in the growth of single-walled carbon nanotubes (SWCNTs), with binary metallic catalysts emerging as an efficient SWCNT synthesis strategy. Among these, iron (Fe), cobalt (Co), and their alloys are particularly effective. However, prior studies have predominantly employed Fe--Co alloy catalysts with fixed atomic ratios as well as unchanged chemical vapor deposition (CVD) conditions, leaving the influence of variable Fe--Co compositions and CVD growth parameters on SWCNT synthesis poorly understood. This study focuses on the role of Fe--Co catalyst ratios, with the aim of elucidating the distinct contributions of Fe and Co atoms in the growth of SWCNTs. By systematically exploring a wide range of Fe--Co ratios and growth conditions, we identified Fe$_{0.75}$Co$_{0.25}$ as a highly efficient binary catalyst at 850~$^\circ$C, primarily forming catalyst clusters with diameters of 2.5--6~nm and yielding SWCNTs with diameters ranging from 0.9--1.1~nm. On the other hand, Fe$_{0}$Co$_{1}$ exhibited higher catalytic activity at 600~$^\circ$C, generating smaller catalyst clusters of 1.5--5~nm and producing SWCNTs with reduced diameters of about 0.6--0.9~nm. Transmission electron microscope (TEM) and energy dispersive X-ray spectroscopy (EDS) analyses reveal that high SWCNT yields correlate with the formation of uniformly sized Fe--Co catalyst particles with surface-segregated Co that optimizes carbon solubility. Molecular dynamics (MD) simulations further corroborate these findings, demonstrating that the structure and melting behavior of Fe$_x$Co$_{1-x}$ clusters depend on cluster size and composition.

cond-mat.mtrl-sci↗

arXiv2025

A Collaborative Framework Integrating Large Language Model and Chemical Fragment Space: Mutual Inspiration for Lead Design

Combinatorial optimization algorithm is essential in computer-aided drug design by progressively exploring chemical space to design lead compounds with high affinity to target protein. However current methods face inherent challenges in integrating domain knowledge, limiting their performance in identifying lead compounds with novel and valid binding mode. Here, we propose AutoLeadDesign, a lead compounds design framework that inspires extensive domain knowledge encoded in large language models with chemical fragments to progressively implement efficient exploration of vast chemical space. The comprehensive experiments indicate that AutoLeadDesign outperforms baseline methods. Significantly, empirical lead design campaigns targeting two clinically relevant targets (PRMT5 and SARS-CoV-2 PLpro) demonstrate AutoLeadDesign's competence in de novo generation of lead compounds achieving expert-competitive design efficacy. Structural analysis further confirms their mechanism-validated inhibitory patterns. By tracing the process of design, we find that AutoLeadDesign shares analogous mechanisms with fragment-based drug design which traditionally rely on the expert decision-making, further revealing why it works. Overall, AutoLeadDesign offers an efficient approach for lead compounds design, suggesting its potential utility in drug design.

q-bio.BM↗

arXiv2025

UniCode$^2$: Cascaded Large-scale Codebooks for Unified Multimodal Understanding and Generation

Unified multimodal large language models (MLLMs) have shown promise in jointly advancing multimodal understanding and generation, with visual codebooks discretizing images into tokens for autoregressive modeling. Existing codebook-based methods either rely on small vocabularies (~16K entries) that lack fine-grained semantics or naively scale up, resulting in low token utilization and unstable training. We propose UniCode$^2$, a cascaded codebook framework enabling large-scale, semantically aligned, and stable visual tokenization. By clustering millions of SigLIP sequence embeddings, we build a 500K-entry codebook that preserves vision-language alignment while expanding capacity. Stability is ensured via a cascaded design: a frozen codebook anchors the embedding space, and a trainable codebook refines task-specific semantics. This decoupling promotes high utilization and robust learning. Moreover, the alignment of our visual tokens with textual semantics enables seamless integration with pretrained diffusion decoders, supporting high-quality visual synthesis with minimal adaptation. UniCode^2 delivers strong performance across diverse benchmarks, demonstrating the viability of scaling visual token spaces without sacrificing stability, semantics, or modularity.

cs.CV↗

arXiv2025

Real-space titration and manipulation of particle-like correlated electrons in doped Mott insulator

The localized (particle-like) correlated electrons deserve particular attention as they govern various exotic quantum phenomena, such as quantum spin liquids, Wigner crystals, and Mott insulators in correlated systems. However, direct observation and manipulation of these particle-like electrons at the atomic or single-electron scale remain highly challenging. Here, we successfully realize and directly visualize particle-like correlated electrons in 1T-TaS2 through hole doping. The potential-dependent local electronic structure of single particle-like electron is revealed by angle-resolved photoemission spectroscopy (ARPES), scanning tunneling spectroscopy (STS) combined with theoretical calculations. The complex correlated interactions including nearest-neighbor attractive interactions and many-body repulsive interactions are further demonstrated and discussed based on the spatial distribution of particle-like electrons. Furthermore, the tentative manipulation of the particle-like electrons is successfully achieved by the energy-excitation mechanism. Our results not only provide profound insights into particle-like electrons in correlated systems, but also establish a versatile platform for designing and controlling quantum states at the atomic scale.

cond-mat.str-el↗

arXiv2025

Numerical Analysis of the Parallel Orbital-Updating Approach for Eigenvalue Problems

The parallel orbital-updating approach is an orbital/eigenfunction iteration based approach for solving eigenvalue problems when many eigenpairs are required. It has been proven to be efficient, for instance, in electronic structure calculations. In this paper, based on the investigation of a quasi-orthogonality, we present the numerical analysis of the parallel orbital-updating approach for linear eigenvalue problems, including convergence and error estimates of the numerical approximations.

math.NA↗

arXiv2025

Enhanced knowledge retention through MedScrab: an interactive mobile game

Noncompliance with medication regimens poses an immense challenge in the management of chronic diseases, often resulting in exacerbated health complications and recurrent hospital admissions. Addressing this gap, our team designed an innovative mobile game aimed at bolstering medication adherence and information retention within the general population. Employing Amazon Mechanical Turk, participants were enlisted and allocated into two cohorts: one engaged with our mobile game and the other perused an informational pamphlet about medication. Both cohorts underwent a pre-intervention quiz, followed by their respective interventions, and concluded with a post-intervention quiz. Primary outcome measures included the difference in quiz scores and the game play duration. The investigation encompassed 243 participants with homogenous baseline attributes. Participants interacting with the mobile game depicted a significant enhancement in their post-intervention scores compared to the pre-intervention scores. We observed a notable correlation of 0.346 (p<0.001) with a robust medium effect size of 0.641 (0.503 - 0.779). Although the duration of game play and post-intervention scores didn't exhibit a direct correlation, a tendency towards superior post-intervention scores was evident among participants who dedicated more time to the game. The interactive mobile game we developed exhibits potential as an engaging instrument for empowering patients and caregivers. Providing critical medication information and the potential side effects in a manner that increases retention would thereby mitigate medication noncompliance. Future research endeavors should focus on optimizing and broadening the application of such mobile interfaces to fortify public health initiatives.

cs.HC↗

arXiv2025

COEF-VQ: Cost-Efficient Video Quality Understanding through a Cascaded Multimodal LLM Framework

Recently, with the emergence of recent Multimodal Large Language Model (MLLM) technology, it has become possible to exploit its video understanding capability on different classification tasks. In practice, we face the difficulty of huge requirements for GPU resource if we need to deploy MLLMs online. In this paper, we propose COEF-VQ, a novel cascaded MLLM framework designed to enhance video quality understanding on the short-video platform while optimizing computational efficiency. Our approach integrates an entropy-based pre-filtering stage, where a lightweight model assesses uncertainty and selectively filters cases before passing them to the more computationally intensive MLLM for final evaluation. By prioritizing high-uncertainty samples for deeper analysis, our framework significantly reduces GPU usage while maintaining the strong classification performance of a full MLLM deployment. To demonstrate the effectiveness of COEF-VQ, we deploy this new framework onto the video management platform (VMP) at the short-video platform, and perform a series of detailed experiments on two in-house tasks related to video quality understanding. We show that COEF-VQ leads to substantial performance gains from the offline evaluation in these two tasks and effectively enhances platform safety with limit resource consumption, significantly reducing inappropriate content video view rate by 9.9% in a online A/B test without affecting engagement. Post-launch monitoring confirmed sustained improvements, validating its real-world impact.

cs.CV↗

arXiv2025

Kwai Keye-VL Technical Report

While Multimodal Large Language Models (MLLMs) demonstrate remarkable capabilities on static images, they often fall short in comprehending dynamic, information-dense short-form videos, a dominant medium in today's digital landscape. To bridge this gap, we introduce \textbf{Kwai Keye-VL}, an 8-billion-parameter multimodal foundation model engineered for leading-edge performance in short-video understanding while maintaining robust general-purpose vision-language abilities. The development of Keye-VL rests on two core pillars: a massive, high-quality dataset exceeding 600 billion tokens with a strong emphasis on video, and an innovative training recipe. This recipe features a four-stage pre-training process for solid vision-language alignment, followed by a meticulous two-phase post-training process. The first post-training stage enhances foundational capabilities like instruction following, while the second phase focuses on stimulating advanced reasoning. In this second phase, a key innovation is our five-mode ``cold-start'' data mixture, which includes ``thinking'', ``non-thinking'', ``auto-think'', ``think with image'', and high-quality video data. This mixture teaches the model to decide when and how to reason. Subsequent reinforcement learning (RL) and alignment steps further enhance these reasoning capabilities and correct abnormal model behaviors, such as repetitive outputs. To validate our approach, we conduct extensive evaluations, showing that Keye-VL achieves state-of-the-art results on public video benchmarks and remains highly competitive on general image-based tasks (Figure 1). Furthermore, we develop and release the \textbf{KC-MMBench}, a new benchmark tailored for real-world short-video scenarios, where Keye-VL shows a significant advantage.

cs.CV↗

arXiv2025

ASR-enhanced Multimodal Representation Learning for Cross-Domain Product Retrieval

E-commerce is increasingly multimedia-enriched, with products exhibited in a broad-domain manner as images, short videos, or live stream promotions. A unified and vectorized cross-domain production representation is essential. Due to large intra-product variance and high inter-product similarity in the broad-domain scenario, a visual-only representation is inadequate. While Automatic Speech Recognition (ASR) text derived from the short or live-stream videos is readily accessible, how to de-noise the excessively noisy text for multimodal representation learning is mostly untouched. We propose ASR-enhanced Multimodal Product Representation Learning (AMPere). In order to extract product-specific information from the raw ASR text, AMPere uses an easy-to-implement LLM-based ASR text summarizer. The LLM-summarized text, together with visual data, is then fed into a multi-branch network to generate compact multimodal embeddings. Extensive experiments on a large-scale tri-domain dataset verify the effectiveness of AMPere in obtaining a unified multimodal product representation that clearly improves cross-domain product retrieval.

cs.MM↗

arXiv2025

Efficient Extreme Operating Condition Search for Online Relay Setting Calculation in Renewable Power Systems Based on Parallel Graph Neural Network

The Extreme Operating Conditions Search (EOCS) problem is one of the key problems in relay setting calculation, which is used to ensure that the setting values of protection relays can adapt to the changing operating conditions of power systems over a period of time after deployment. The high penetration of renewable energy and the wide application of inverter-based resources make the operating conditions of renewable power systems more volatile, which urges the adoption of the online relay setting calculation strategy. However, the computation speed of existing EOCS methods based on local enumeration, heuristic algorithms, and mathematical programming cannot meet the efficiency requirement of online relay setting calculation. To reduce the time overhead, this paper, for the first time, proposes an efficient deep learning-based EOCS method suitable for online relay setting calculation. First, the power system information is formulated as four layers, i.e., a component parameter layer, a topological connection layer, an electrical distance layer, and a graph distance layer, which are fed into a parallel graph neural network (PGNN) model for feature extraction. Then, the four feature layers corresponding to each node are spliced and stretched, and then fed into the decision network to predict the extreme operating condition of the system. Finally, the proposed PGNN method is validated on the modified IEEE 39-bus and 118-bus test systems, where some of the synchronous generators are replaced by renewable generation units. The nonlinear fault characteristics of renewables are fully considered when computing fault currents. The experiment results show that the proposed PGNN method achieves higher accuracy than the existing methods in solving the EOCS problem. Meanwhile, it also provides greater improvements in online computation time.

cs.LG↗

arXiv2025

AviationLLM: An LLM-based Knowledge System for Aviation Training

Aviation training is a core link in ensuring flight safety, improving industry efficiency and promoting sustainable development. It not only involves flight simulation but also requires the learning of a great deal of professional aviation theory knowledge. In the existing training system, the knowledge is mainly imparted by the the instructors. However, the number of instructors is limited and the professional answers obtained from the Internet are not accurate enough, resulting in low training efficiency. To address this, we introduced LLM, but the basic pre-trained model cannot provide accurate answers to professional fields, so we fine-tuned it. Traditional Supervised Fine-Tuning (SFT) risk generating superficially plausible but factually incorrect responses due to insufficient data coverage. To address this, we employ Direct Preference Optimization(DPO). This paper proposes Retrieval-Augmented LLM Alignment via Direct Preference Optimization(RALA-DPO). We select open source pre-trained LLM Qwen and adapt it to aviation theory training through DPO-based domain alignment. Simultaneously, to mitigate hallucinations caused by training data biases, knowledge obsolescence, or domain knowledge gaps, we implement Retrieval-Augmented Generation(RAG) technology that combines generative and retrieval models. RALA-DPO effectively retrieves relevant information from external knowledge bases and delivers precise and high-quality responses through the generative model. Experimental results demonstrate that RALA-DPO can improve accuracy in response to professional aviation knowledge. With integrated RAG mechanisms, this system can further improve the accuracy of answers and achieve zero-cost knowledge updates simultaneously.

cs.AI↗

arXiv2025

EKPC: Elastic Knowledge Preservation and Compensation for Class-Incremental Learning

Class-Incremental Learning (CIL) aims to enable AI models to continuously learn from sequentially arriving data of different classes over time while retaining previously acquired knowledge. Recently, Parameter-Efficient Fine-Tuning (PEFT) methods, like prompt pool-based approaches and adapter tuning, have shown great attraction in CIL. However, these methods either introduce additional parameters that increase memory usage, or rely on rigid regularization techniques which reduce forgetting but compromise model flexibility. To overcome these limitations, we propose the Elastic Knowledge Preservation and Compensation (EKPC) method, integrating Importance-aware Parameter Regularization (IPR) and Trainable Semantic Drift Compensation (TSDC) for CIL. Specifically, the IPR method assesses the sensitivity of network parameters to prior tasks using a novel parameter-importance algorithm. It then selectively constrains updates within the shared adapter according to these importance values, thereby preserving previously acquired knowledge while maintaining the model's flexibility. However, it still exhibits slight semantic differences in previous knowledge to accommodate new incremental tasks, leading to decision boundaries confusion in classifier. To eliminate this confusion, TSDC trains a unified classifier by compensating prototypes with trainable semantic drift. Extensive experiments on five CIL benchmarks demonstrate the effectiveness of the proposed method, showing superior performances to existing state-of-the-art methods.

cs.CV↗

arXiv2025

VRD-IU: Lessons from Visually Rich Document Intelligence and Understanding

Visually Rich Document Understanding (VRDU) has emerged as a critical field in document intelligence, enabling automated extraction of key information from complex documents across domains such as medical, financial, and educational applications. However, form-like documents pose unique challenges due to their complex layouts, multi-stakeholder involvement, and high structural variability. Addressing these issues, the VRD-IU Competition was introduced, focusing on extracting and localizing key information from multi-format forms within the Form-NLU dataset, which includes digital, printed, and handwritten documents. This paper presents insights from the competition, which featured two tracks: Track A, emphasizing entity-based key information retrieval, and Track B, targeting end-to-end key information localization from raw document images. With over 20 participating teams, the competition showcased various state-of-the-art methodologies, including hierarchical decomposition, transformer-based retrieval, multimodal feature fusion, and advanced object detection techniques. The top-performing models set new benchmarks in VRDU, providing valuable insights into document intelligence.

cs.CV↗

arXiv2025

A Unified Solution to Diverse Heterogeneities in One-shot Federated Learning

One-Shot Federated Learning (OSFL) restricts communication between the server and clients to a single round, significantly reducing communication costs and minimizing privacy leakage risks compared to traditional Federated Learning (FL), which requires multiple rounds of communication. However, existing OSFL frameworks remain vulnerable to distributional heterogeneity, as they primarily focus on model heterogeneity while neglecting data heterogeneity. To bridge this gap, we propose FedHydra, a unified, data-free, OSFL framework designed to effectively address both model and data heterogeneity. Unlike existing OSFL approaches, FedHydra introduces a novel two-stage learning mechanism. Specifically, it incorporates model stratification and heterogeneity-aware stratified aggregation to mitigate the challenges posed by both model and data heterogeneity. By this design, the data and model heterogeneity issues are simultaneously monitored from different aspects during learning. Consequently, FedHydra can effectively mitigate both issues by minimizing their inherent conflicts. We compared FedHydra with five SOTA baselines on four benchmark datasets. Experimental results show that our method outperforms the previous OSFL methods in both homogeneous and heterogeneous settings. The code is available at https://github.com/Jun-B0518/FedHydra.

cs.DC↗

arXiv2025

MIDAS: Multi-level Intent, Domain, And Slot Knowledge Distillation for Multi-turn NLU

Although Large Language Models (LLMs) can generate coherent text, they often struggle to recognise user intent behind queries. In contrast, Natural Language Understanding (NLU) models interpret the purpose and key information of user input for responsive interactions. Existing NLU models typically map utterances to a dual-level semantic frame, involving sentence-level intent (SI) and word-level slot (WS) labels. However, real-life conversations primarily consist of multi-turn dialogues, requiring the interpretation of complex and extended exchanges. Researchers encounter challenges in addressing all facets of multi-turn dialogue using a unified NLU model. This paper introduces MIDAS, a novel approach leveraging multi-level intent, domain, and slot knowledge distillation for multi-turn NLU. We construct distinct teachers for SI detection, WS filling, and conversation-level domain (CD) classification, each fine-tuned for specific knowledge. A multi-teacher loss is proposed to facilitate the integration of these teachers, guiding a student model in multi-turn dialogue tasks. Results demonstrate the efficacy of our model in improving multi-turn conversation understanding, showcasing the potential for advancements in NLU through multi-level dialogue knowledge distillation. Our implementation is open-sourced on https://github.com/adlnlp/Midas.

cs.CL↗