arXiv subjects

Li Yang

Publications and source records attributed to Li Yang.

At least 55 records · Page 3Linked to original sources

arXiv2026

Divide-then-Diagnose: Weaving Clinician-Inspired Contexts for Ultra-Long Capsule Endoscopy Videos

Capsule endoscopy (CE) enables non-invasive gastrointestinal screening, but current CE research remains largely limited to frame-level classification and detection, leaving video-level analysis underexplored. To bridge this gap, we introduce and formally define a new task, diagnosis-driven CE video summarization, which requires extracting key evidence frames that covers clinically meaningful findings and making accurate diagnoses from those evidence frames. This setting is challenging because diagnostically relevant events are extremely sparse and can be overwhelmed by tens of thousands of redundant normal frames, while individual observations are often ambiguous due to motion blur, debris, specular highlights, and rapid viewpoint changes. To facilitate research in this direction, we introduce VideoCAP, the first CE dataset with diagnosis-driven annotations derived from real clinical reports. VideoCAP comprises 240 full-length videos and provides realistic supervision for both key evidence frame extraction and diagnosis. To address this task, we further propose DiCE, a clinician-inspired framework that mirrors the standard CE reading workflow. DiCE first performs efficient candidate screening over the raw video, then uses a Context Weaver to organize candidates into coherent diagnostic contexts that preserve distinct lesion events, and an Evidence Converger to aggregate multi-frame evidence within each context into robust clip-level judgments. Experiments show that DiCE consistently outperforms state-of-the-art methods, producing concise and clinically reliable diagnostic summaries. These results highlight diagnosis-driven contextual reasoning as a promising paradigm for ultra-long CE video summarization.

cs.CV↗

arXiv2026

NTIRE 2026 Challenge on Video Saliency Prediction: Methods and Results

This paper presents an overview of the NTIRE 2026 Challenge on Video Saliency Prediction. The goal of the challenge participants was to develop automatic saliency map prediction methods for the provided video sequences. The novel dataset of 2,000 diverse videos with an open license was prepared for this challenge. The fixations and corresponding saliency maps were collected using crowdsourced mouse tracking and contain viewing data from over 5,000 assessors. Evaluation was performed on a subset of 800 test videos using generally accepted quality metrics. The challenge attracted over 20 teams making submissions, and 7 teams passed the final phase with code review. All data used in this challenge is made publicly available - https://github.com/msu-video-group/NTIRE26_Saliency_Prediction.

cs.CV↗

arXiv2026

TempR1: Improving Temporal Understanding of MLLMs via Temporal-Aware Multi-Task Reinforcement Learning

Enhancing the temporal understanding of Multimodal Large Language Models (MLLMs) is essential for advancing long-form video analysis, enabling tasks such as temporal localization, action detection, and time-sensitive question answering. While reinforcement learning (RL) has recently been explored for improving temporal reasoning, existing approaches are often confined to limited task types and data, restricting their generalization across diverse temporal understanding scenarios. To address this challenge, we present TempR1, a temporal-aware multi-task reinforcement learning framework that systematically strengthens MLLMs' temporal comprehension. We curate a multi-task corpus that exposes the model to diverse temporal structures and semantics, and build upon the Group Relative Policy Optimization (GRPO) algorithm to achieve stable and effective cross-task optimization. Specifically, we categorize temporal tasks into three correspondence types between predicted intervals and ground-truth instances, and design tailored localization rewards for each, enabling TempR1 to capture fine-grained temporal dependencies and adapt to different temporal patterns. Extensive experiments demonstrate that TempR1 attains state-of-the-art performance across multiple benchmarks. Moreover, its joint optimization over complementary tasks yields a strong synergistic effect, enhancing both generalization and single-task performance, establishing a scalable and principled paradigm for temporal reasoning in MLLMs.

cs.CV↗

arXiv2026

Building an Affordable Self-Driving Lab: Practical Machine Learning Experiments for Physics Education Using Internet-of-Things

Machine learning (ML) is transforming modern physics research, but practical, hands-on experience with ML techniques remains limited due to cost and complexity barriers. To address this gap, we introduce an affordable, autonomous, Internet-of-Things (IoT)-enabled experimental platform designed specifically for applied physics education. Utilizing an Arduino microcontroller, a customizable multi-wavelength light emitting diode (LED) array, and photosensors, our setup generates diverse, real-time optical datasets ideal for training and evaluating foundational ML algorithms, including traversal methods, Bayesian inference, and deep learning. The platform facilitates a closed-loop, self-driving experimental workflow, encompassing automated data collection, preprocessing, model training, and validation. Through systematic performance comparisons, we demonstrate the superior ability of deep learning to capture complex nonlinear relationships compared to traversal and Bayesian methods. At approximately $60, this open-source IoT platform provides an accessible, practical pathway for students to master advanced ML concepts, promoting deeper conceptual insights and essential technical skills required for the next generation of physicists and engineers.

physics.ed-ph↗

arXiv2026

LiveGesture Streamable Co-Speech Gesture Generation Model

We propose LiveGesture, the first fully streamable, speech-driven full-body gesture generation framework that operates with zero look-ahead and supports arbitrary sequence length. Unlike existing co-speech gesture methods, which are designed for offline generation and either treat body regions independently or entangle all joints within a single model, LiveGesture is built from the ground up for causal, region-coordinated motion generation. LiveGesture consists of two main modules: the Streamable Vector Quantized Motion Tokenizer (SVQ) and the Hierarchical Autoregressive Transformer (HAR). The SVQ tokenizer converts the motion sequence of each body region into causal, discrete motion tokens, enabling real-time, streamable token decoding. On top of SVQ, HAR employs region-expert autoregressive (xAR) transformers to model expressive, fine-grained motion dynamics for each body region. A causal spatio-temporal fusion module (xAR Fusion) then captures and integrates correlated motion dynamics across regions. Both xAR and xAR Fusion are conditioned on live, continuously arriving audio signals encoded by a streamable causal audio encoder. To enhance robustness under streaming noise and prediction errors, we introduce autoregressive masking training, which leverages uncertainty-guided token masking and random region masking to expose the model to imperfect, partially erroneous histories during training. Experiments on the BEAT2 dataset demonstrate that LiveGesture produces coherent, diverse, and beat-synchronous full-body gestures in real time, matching or surpassing state-of-the-art offline methods under true zero look-ahead conditions.

cs.CV↗

arXiv2026

Towards Resilient Intrusion Detection in CubeSats: Challenges, TinyML Solutions, and Future Directions

CubeSats have revolutionized access to space by providing affordable and accessible platforms for research and education. However, their reliance on Commercial Off-The-Shelf (COTS) components and open-source software has introduced significant cybersecurity vulnerabilities. Ensuring the cybersecurity of CubeSats is vital as they play increasingly important roles in space missions. Traditional security measures, such as intrusion detection systems (IDS), are impractical for CubeSats due to resource constraints and unique operational environments. This paper provides an in-depth review of current cybersecurity practices for CubeSats, highlighting limitations and identifying gaps in existing methods. Additionally, it explores non-cyber anomaly detection techniques that offer insights into adaptable algorithms and deployment strategies suitable for CubeSat constraints. Open research problems are identified, including the need for resource-efficient intrusion detection mechanisms, evaluation of IDS solutions under realistic mission scenarios, development of autonomous response systems, and creation of cybersecurity frameworks. The addition of TinyML into CubeSat systems is explored as a promising solution to address these challenges, offering resource-efficient, real-time intrusion detection capabilities. Future research directions are proposed, such as integrating cybersecurity with health monitoring systems, and fostering collaboration between cybersecurity researchers and space domain experts.

cs.CR↗

arXiv2026

Twist-induced Out-of-plane Ferroelectricity in Bilayer Hafnia

Ferroelectric HfO2 is a promising candidate for next-generation memory devices due to its CMOS compatibility and ability to retain polarization at nanometer scales. However, the polar orthorhombic phase (Pca2_1) responsible for ferroelectricity is metastable and requires extrinsic stabilization, which makes it challenging for integration with silicon. We predict that bilayer 1T-HfO2 can exhibit robust and switchable out-of-plane (OOP) polarization arising from stacking-induced symmetry breaking. Using first-principles density functional theory, we predict that monolayer 1T-HfO2 can be cleaved from the (111) surface of cubic hafnia, and the monolayer is dynamically stable. When two aligned monolayers are twisted to form a moiré superlattice, it breaks the interlayer symmetry and allows the emergence of bistable OOP polarization. At a twist angle of 7.34o, the system exhibits a net polarization of ~16 μC/cm2. This sizeable polarization is due to the large polar displacements concentrated in AB stacking domains. Importantly, this polarization can be reversibly switched via interlayer sliding with a low energy barrier (~8 meV/formula unit) and comparable low coercive field (~0.2 V/nm), offering electric-field tunability. These findings establish twisted bilayer 1T-HfO2 as a scalable and robust 2D ferroelectric platform, enabling new pathways for integrating ferroelectric functionality into atomically thin memory and logic devices.

cond-mat.mtrl-sci↗

arXiv2026

SQL-Commenter: Aligning Large Language Models for SQL Comment Generation with Direct Preference Optimization

SQL query comprehension is a significant challenge due to complex syntax, diverse join types, and deep nesting. Many queries lack adequate comments, severely hindering code readability, maintainability, and knowledge transfer. Automated SQL comment generation faces two main challenges: limited datasets that inadequately represent complex real-world queries, and Large Language Models' (LLMs) insufficient understanding of SQL-specific semantics. Our empirical analysis shows that even after continual pre-training and supervised fine-tuning, LLMs struggle with complex SQL semantics, yielding inaccurate comments. To address this, we propose SQL-Commenter, an advanced method based on LLaMA-3.1-8B. We first construct a comprehensive dataset of complex SQL queries with expert-verified comments. Next, we perform continual pre-training on a large SQL corpus to enhance the LLM's syntax and semantic understanding, followed by supervised fine-tuning. Finally, we introduce Direct Preference Optimization (DPO) using human feedback. SQL-Commenter utilizes a preference-based loss function to favor preferred outputs, enhancing fine-grained semantic learning and context-dependent quality assessment. Evaluated on the Spider and Bird benchmarks, SQL-Commenter significantly outperforms state-of-the-art baselines. On average, it surpasses the strongest baseline (Qwen3-14B) by 9.29, 4.99, and 13.23 percentage points on BLEU-4, METEOR, and ROUGE-L, respectively. Moreover, human evaluation demonstrates the superior quality of comments generated by SQL-Commenter in terms of correctness, completeness, and naturalness.

cs.SE↗

arXiv2026

More Than Memory Savings: Zeroth-Order Optimization Mitigates Forgetting in Continual Learning

Zeroth-order (ZO) optimization has gained attention as a memory-efficient alternative to first-order (FO) methods, particularly in settings where gradient computation is expensive or even impractical. Beyond its memory efficiency, in this work, we investigate ZO optimization for continual learning (CL) as a novel approach to address the plasticity-stability-efficiency trilemma. Through theoretical analysis and empirical evidence, we show that ZO optimization naturally leads to flatter loss landscapes, which in turn reduce forgetting in CL. However, this stability comes at a cost of plasticity: due to its imprecise gradient estimates and slower convergence, ZO optimization tends to be less effective than FO in acquiring new task-specific knowledge, particularly under constrained training budgets. To better understand this trade-off, we conduct a holistic evaluation of ZO optimization applied to various existing CL methods. Our findings reveal that ZO optimization enhances stability but often undermines plasticity, particularly when used with learnable classifiers. Motivated by this insight, we propose ZO-FC, a simple but effective approach that applies ZO optimization to a single adapter-based PEFT module with FO optimized classifier. This design leverages the stability benefits of ZO while preserving the adaptability of FO updates with negligible memory overhead. Experiments demonstrate that ZO-FC achieves an effective balance between stability and plasticity, offering a practical and memory-efficient solution for on-device CL.

cs.LG↗

arXiv2026

Rethinking Continual Learning with Progressive Neural Collapse

Continual Learning (CL) seeks to build an agent that can continuously learn a sequence of tasks, where a key challenge, namely Catastrophic Forgetting, persists due to the potential knowledge interference among different tasks. On the other hand, deep neural networks (DNNs) are shown to converge to a terminal state termed Neural Collapse during training, where all class prototypes geometrically form a static simplex equiangular tight frame (ETF). These maximally and equally separated class prototypes make the ETF an ideal target for model learning in CL to mitigate knowledge interference. Thus inspired, several studies have emerged very recently to leverage a fixed global ETF in CL, which however suffers from key drawbacks, such as impracticability and limited performance.To address these challenges and fully unlock the potential of ETF in CL, we propose Progressive Neural Collapse (ProNC), a novel framework that completely removes the need of a fixed global ETF in CL. Specifically, ProNC progressively expands the ETF target in a principled way by adding new class prototypes as vertices for new tasks, ensuring maximal separability across all encountered classes with minimal shifts from the previous ETF. We next develop a new CL framework by plugging ProNC into commonly used CL algorithm designs, where distillation is further leveraged to balance between target shifting for old classes and target aligning for new classes. Extensive experiments show that our approach significantly outperforms related baselines while maintaining superior flexibility, simplicity, and efficiency. Our code is available at https://github.com/Continue-Edge-AI-Lab/ProNC

cs.LG↗

arXiv2026

A multitask framework for automated interpretation of multi-frame right upper quadrant ultrasound in clinical decision support

Ultrasound is a cornerstone of emergency and hepatobiliary imaging, yet its interpretation remains highly operator-dependent and time-sensitive. Here, we present a multitask vision-language agent (VLM) developed to assist with comprehensive right upper quadrant (RUQ) ultrasound interpretation across the full diagnostic workflow. The system was trained on a large, multi-center dataset comprising a primary cohort from Johns Hopkins Medical Institutions (9,189 cases, 594,099 images) and externally validated on cohorts from Stanford University (108 cases, 3,240 images) and a major Chinese medical center (257 cases, 3,178 images). Built on the Qwen2.5-VL-7B architecture, the agent integrates frame-level visual understanding with report-grounded language reasoning to perform three tasks: (i) classification of 18 hepatobiliary and gallbladder conditions, (ii) generation of clinically coherent diagnostic reports, and (iii) surgical decision support based on ultrasound findings and clinical data. The model achieved high diagnostic accuracy across all tasks, generated reports that were indistinguishable from expert-written versions in blinded evaluations, and demonstrated superior factual accuracy and information density on content-based metrics. The agent further identified patients requiring cholecystectomy with high precision, supporting real-time decision-making. These results highlight the potential of generalist vision-language models to improve diagnostic consistency, reporting efficiency, and surgical triage in real-world ultrasound practice.

eess.IV↗

arXiv2025

Prior-AttUNet: Retinal OCT Fluid Segmentation Based on Normal Anatomical Priors and Attention Gating

Accurate segmentation of macular edema, a hallmark pathological feature in vision-threatening conditions such as age-related macular degeneration and diabetic macular edema, is essential for clinical diagnosis and management. To overcome the challenges of segmenting fluid regions in optical coherence tomography (OCT) images-notably ambiguous boundaries and cross-device heterogeneity-this study introduces Prior-AttUNet, a segmentation model augmented with generative anatomical priors. The framework adopts a hybrid dual-path architecture that integrates a generative prior pathway with a segmentation network. A variational autoencoder supplies multi-scale normative anatomical priors, while the segmentation backbone incorporates densely connected blocks and spatial pyramid pooling modules to capture richer contextual information. Additionally, a novel triple-attention mechanism, guided by anatomical priors, dynamically modulates feature importance across decoding stages, substantially enhancing boundary delineation. Evaluated on the public RETOUCH benchmark, Prior-AttUNet achieves excellent performance across three OCT imaging devices (Cirrus, Spectralis, and Topcon), with mean Dice similarity coefficients of 93.93%, 95.18%, and 93.47%, respectively. The model maintains a low computational cost of 0.37 TFLOPs, striking an effective balance between segmentation precision and inference efficiency. These results demonstrate its potential as a reliable tool for automated clinical analysis.

cs.CV↗

arXiv2025

Itinerant Magnetism in Twisted Bilayer WSe$_2$ and MoTe$_2$

Using a self-consistent Hartree-Fock theory, we show that the recently observed ferromagnetism in twisted bilayer WSe$_2$ [Nat. Commun. 16, 1959 (2025)] can be understood as a Stoner-like instability of interaction-renormalized moiré bands. We quantitatively reproduce the observed Lifshitz transition as function of hole filling and applied electric field that marks the boundary between layer-hybridized and layer-polarized regimes. The former supports a ferromagnetic valley-polarized ground state below half-filling, developing a topological charge gap at half-filling for smaller twist angles. At larger twist angles, the system hosts a gapped triangular Néel antiferromagnet. On the other hand, the layer-polarized regime supports a stripe antiferromagnet below half-filling and a wing-shaped multiferroic ground state above half-filling. We map the evolution of these states as a function of filling factor, electric field, twist angle, and interaction strength. Our results demonstrate that long-range exchange in a symmetry-unbroken parent state with strongly renormalized moiré bands provides a broadly applicable framework to understand itinerant magnetism in moiré TMDs.

cond-mat.str-el↗

arXiv2025

Think Before You Prune: Self-Reflective Structured Pruning for Reasoning Language Models

Reasoning LLMs (RLMs) such as OpenAI o1, DeepSeek-R1, and Qwen3 deliver strong multi-step reasoning through chain-of-thought generation, but their large model sizes and lengthy decode-time outputs make them costly to deploy and unsuitable for resource-constrained settings. To reduce computing and memory cost, pruning offers a promising solution by removing unimportant parameters. However, despite their success on standard LLMs, existing pruning methods severely damage RLMs, as even moderate sparsity (e.g., 20%) can collapse accuracy and completely disrupt the model's reasoning coherence. We begin by analyzing why existing pruning pipelines fail on reasoning LLMs and find that their brittleness largely stems from a mismatch between the calibration data, the pruning objective, and the model's decode-time reasoning behavior. Our study further shows that the most reliable calibration signal comes not from human-written labels but from the model's own self-generated reasoning traces, which more accurately reflect its inference distribution. Guided by these insights, we introduce RESP, a self-reflective structured pruning framework that aligns pruning decisions with the model's reasoning dynamics through self-generated calibration, decode-only gradient-based importance estimation, and progressive regeneration that maintains calibration fidelity as sparsity increases. Experiments on Qwen3-8B demonstrate that RESP markedly outperforms existing structured pruning methods on both GSM8K and MathQA, preserving near-dense accuracy at 20-30% sparsity and substantially mitigating performance collapse at higher sparsity levels. At 40% sparsity, RESP attains 81.3% accuracy on GSM8K and 59.6% on MathQA, surpassing the strongest baselines by 66.87% and 47%, respectively.

cs.CL↗

arXiv2025

DCL-SE: Dynamic Curriculum Learning for Spatiotemporal Encoding of Brain Imaging

High-dimensional neuroimaging analyses for clinical diagnosis are often constrained by compromises in spatiotemporal fidelity and by the limited adaptability of large-scale, general-purpose models. To address these challenges, we introduce Dynamic Curriculum Learning for Spatiotemporal Encoding (DCL-SE), an end-to-end framework centered on data-driven spatiotemporal encoding (DaSE). We leverage Approximate Rank Pooling (ARP) to efficiently encode three-dimensional volumetric brain data into information-rich, two-dimensional dynamic representations, and then employ a dynamic curriculum learning strategy, guided by a Dynamic Group Mechanism (DGM), to progressively train the decoder, refining feature extraction from global anatomical structures to fine pathological details. Evaluated across six publicly available datasets, including Alzheimer's disease and brain tumor classification, cerebral artery segmentation, and brain age prediction, DCL-SE consistently outperforms existing methods in accuracy, robustness, and interpretability. These findings underscore the critical importance of compact, task-specific architectures in the era of large-scale pretrained networks.

cs.CV↗

arXiv2025

Toward Autonomous and Efficient Cybersecurity: A Multi-Objective AutoML-based Intrusion Detection System

With increasingly sophisticated cybersecurity threats and rising demand for network automation, autonomous cybersecurity mechanisms are becoming critical for securing modern networks. The rapid expansion of Internet of Things (IoT) systems amplifies these challenges, as resource-constrained IoT devices demand scalable and efficient security solutions. In this work, an innovative Intrusion Detection System (IDS) utilizing Automated Machine Learning (AutoML) and Multi-Objective Optimization (MOO) is proposed for autonomous and optimized cyber-attack detection in modern networking environments. The proposed IDS framework integrates two primary innovative techniques: Optimized Importance and Percentage-based Automated Feature Selection (OIP-AutoFS) and Optimized Performance, Confidence, and Efficiency-based Combined Algorithm Selection and Hyperparameter Optimization (OPCE-CASH). These components optimize feature selection and model learning processes to strike a balance between intrusion detection effectiveness and computational efficiency. This work presents the first IDS framework that integrates all four AutoML stages and employs multi-objective optimization to jointly optimize detection effectiveness, efficiency, and confidence for deployment in resource-constrained systems. Experimental evaluations over two benchmark cybersecurity datasets demonstrate that the proposed MOO-AutoML IDS outperforms state-of-the-art IDSs, establishing a new benchmark for autonomous, efficient, and optimized security for networks. Designed to support IoT and edge environments with resource constraints, the proposed framework is applicable to a variety of autonomous cybersecurity applications across diverse networked environments.

cs.CR↗

arXiv2025

Asymptotics for Reinforced Stochastic Processes on Hierarchical Networks

In this paper, we analyze the asymptotic behavior of a system of interacting reinforced stochastic processes $({\bf Z}_n, {\bf N}_n)_n$ on a directed network of $N$ agents. The system is defined by the coupled dynamics ${\bf Z}_{n+1}=(1-r_{n}){\bf Z}_{n}+r_{n}{\bf X}_{n+1}$ and ${\bf N}_{n+1}=(1-\frac{1}{n+1}){\bf N}_n+\frac{1}{n+1}{\bf X}_{n+1}$, where agent actions $\mathbb{P}(X_{n+1,j}=1\mid{\cal F}_n)=\sum_{h} w_{hj}Z_{nh}$ are governed by a column-normalized adjacency matrix ${\bf W}$, and $r_n \sim cn^{-γ}$ with $γ\in (1/2, 1]$. Existing asymptotic theory has largely been restricted to irreducible and diagonalizable ${\bf W}$. We extend this analysis to the broader and more practical class of reducible and non-diagonalizable matrices ${\bf W}$ possessing a block upper-triangular form, which models hierarchical influence. We first establish synchronization, proving $({\bf Z}^\top_n, {\bf N}^\top_n)^\top \to Z_\infty {\bf 1}$ almost surely, where the distribution of the limit $Z_\infty$ is shown to be determined solely by the internal dynamics of the leading subgroup. Furthermore, we establish a joint central limit theorem for $({\bf Z}_n,{\bf N}_n)_n$, revealing how the spectral properties and Jordan block structure of ${\bf W}$ govern second-order fluctuations. We demonstrate that the convergence rates and the limiting covariance structure exhibit a phase transition dependent on $γ$ and the spectral properties of ${\bf W}$. Crucially, we explicitly characterize how the non-diagonalizability of ${\bf W}$ fundamentally alters the asymptotic covariance and introduces new logarithmic scaling factors in the critical case ($γ=1$). These results provide a probabilistic foundation for statistical inference on such hierarchical network structures.

math.ST↗

arXiv2025

A Multi-Stage Automated Online Network Data Stream Analytics Framework for IIoT Systems

Industry 5.0 aims at maximizing the collaboration between humans and machines. Machines are capable of automating repetitive jobs, while humans handle creative tasks. As a critical component of Industrial Internet of Things (IIoT) systems for service delivery, network data stream analytics often encounter concept drift issues due to dynamic IIoT environments, causing performance degradation and automation difficulties. In this paper, we propose a novel Multi-Stage Automated Network Analytics (MSANA) framework for concept drift adaptation in IIoT systems, consisting of dynamic data pre-processing, the proposed Drift-based Dynamic Feature Selection (DD-FS) method, dynamic model learning & selection, and the proposed Window-based Performance Weighted Probability Averaging Ensemble (W-PWPAE) model. It is a complete automated data stream analytics framework that enables automatic, effective, and efficient data analytics for IIoT systems in Industry 5.0. Experimental results on two public IoT datasets demonstrate that the proposed framework outperforms state-of-the-art methods for IIoT data stream analytics.

cs.LG↗