Yuxi Zhou

Zhihao Liu, Yifan Wu, Jian Lou et al.

Safety alignment for large language models (LLMs) aims to reduce harmful or unsafe behavior while preserving general utility. However, recent findings reveal that alignment effects can be fragile: lightweight post-alignment manipulations, such as parameter noise, activation noise, or quantization, can easily weaken the intended safety behavior. Prior efforts to improve robustness have primarily focused on data curation, modified alignment objectives, and safety-critical parameter identification, leaving the role of the optimizer itself largely unexplored. In this paper, we are the first to study the robustness of safety alignment from the perspective of the base optimizer. This optimizer-centric view naturally points to zeroth-order optimization, which provides a robustness-oriented signal by evaluating safety alignment under perturbations. Based on this insight, we propose a hybrid framework that first performs standard first-order safety alignment and then applies zeroth-order refinement to improve robustness. Both theoretically and empirically, we show that only a few zeroth-order refinement steps can enhance robustness while preserving safety alignment. We further improve the efficiency of zeroth-order refinement by exploiting its inherent perturbation-based evaluations to estimate layer-wise robustness sensitivity, enabling the refinement process to concentrate updates on robustness-critical layers with modest training overhead.

Yuxi Zhou, Zhengbo Zhang, Jingyu Pan et al.

Human action recognition is pivotal in computer vision, with applications ranging from surveillance to human-robot interaction. Despite the effectiveness of supervised skeleton-based methods, their reliance on exhaustive annotation limits generalization to novel actions. Zero-Shot Skeleton Action Recognition (ZSAR) emerges as a promising paradigm, yet it faces challenges due to the spectral bias of diffusion models, which oversmooth high-frequency dynamics. Here, we propose Frequency-Aware Diffusion for Skeleton-Text Matching (FDSM), integrating a Semantic-Guided Spectral Residual Module, a Timestep-Adaptive Spectral Loss, and Curriculum-based Semantic Abstraction to address these challenges. Our approach effectively recovers fine-grained motion details, achieving state-of-the-art performance on NTU RGB+D, PKU-MMD, and Kinetics-skeleton datasets. Code has been made available at https://github.com/yuzhi535/FDSM. Project homepage: https://yuzhi535.github.io/FDSM.github.io/

Zeyu Wang, Huiying Zhao, Peng Ren et al.

Sepsis is a leading cause of death in the ICU. It is a disease requiring complex interventions in a short period of time, but its optimal treatment strategy remains uncertain. Evidence suggests that the practices of currently used treatment strategies are problematic and may cause harm to patients. To address this decision problem, we propose a new medical decision model based on historical data to help clinicians recommend the best reference option for real-time treatment. Our model combines offline reinforcement learning and deep reinforcement learning to solve the problem of traditional reinforcement learning in the medical field due to the inability to interact with the environment, while enabling our model to make decisions in a continuous state-action space. We demonstrate that, on average, the treatments recommended by the model are more valuable and reliable than those recommended by clinicians. In a large validation dataset, we find out that the patients whose actual doses from clinicians matched the decisions made by AI has the lowest mortality rates. Our model provides personalized and clinically interpretable treatment decisions for sepsis to improve patient care.

Donglin Xie, Qingshuo Zhao, Jingyu Wang et al.

Sleep disturbances are tightly linked to cardiovascular risk, yet polysomnography (PSG)-the clinical reference standard-remains resource-intensive and poorly suited for multi-night, home-based, and large-scale screening. Single-lead electrocardiography (ECG), already ubiquitous in Holter and patch-based devices, enables comfortable long-term acquisition and encodes sleep-relevant physiology through autonomic modulation and cardiorespiratory coupling. Here, we present a proof-of-concept Holter-to-Sleep framework that, using single-lead ECG as the sole input, jointly supports overnight sleep phenotyping and Holter-grade cardiac phenotyping within the same recording, and further provides an explicit analytic pathway for scalable cardio-sleep association studies. The framework is developed and validated on a pooled multi-center PSG sample of 10,439 studies spanning four public cohorts, with independent external evaluation to assess cross-cohort generalizability, and additional real-world feasibility assessment using overnight patch-ECG recordings via objective-subjective consistency analysis. This integrated design enables robust extraction of clinically meaningful overnight sleep phenotypes under heterogeneous populations and acquisition conditions, and facilitates systematic linkage between ECG-derived sleep metrics and arrhythmia-related Holter phenotypes. Collectively, the Holter-to-Sleep paradigm offers a practical foundation for low-burden, home-deployable, and scalable cardio-sleep monitoring and research beyond traditional PSG-centric workflows.

Jiaze Wang, Qinghao Zhao, Zizheng Chen et al.

Traditional diagnosis of aortic valve disease relies on echocardiography, but its cost and required expertise limit its use in large-scale early screening. Photoplethysmography (PPG) has emerged as a promising screening modality due to its widespread availability in wearable devices and its ability to reflect underlying hemodynamic dynamics. However, the extreme scarcity of gold-standard labeled PPG data severely constrains the effectiveness of data-driven approaches. To address this challenge, we propose and validate a new paradigm, Physiology-Guided Self-Supervised Learning (PG-SSL), aimed at unlocking the value of large-scale unlabeled PPG data for efficient screening of Aortic Stenosis (AS) and Aortic Regurgitation (AR). Using over 170,000 unlabeled PPG samples from the UK Biobank, we formalize clinical knowledge into a set of PPG morphological phenotypes and construct a pulse pattern recognition proxy task for self-supervised pre-training. A dual-branch, gated-fusion architecture is then employed for efficient fine-tuning on a small labeled subset. The proposed PG-SSL framework achieves AUCs of 0.765 and 0.776 for AS and AR screening, respectively, significantly outperforming supervised baselines trained on limited labeled data. Multivariable analysis further validates the model output as an independent digital biomarker with sustained prognostic value after adjustment for standard clinical risk factors. This study demonstrates that PG-SSL provides an effective, domain knowledge-driven solution to label scarcity in medical artificial intelligence and shows strong potential for enabling low-cost, large-scale early screening of aortic valve disease.

Yongchao Long, Chao Yang, Gongzheng Tang et al.

Privacy-preserving medical decision support for kidney disease requires localized deployment of large language models (LLMs) while maintaining clinical reasoning capabilities. Current solutions face three challenges: 1) Cloud-based LLMs pose data security risks; 2) Local model deployment demands technical expertise; 3) General LLMs lack mechanisms to integrate medical knowledge. Retrieval-augmented systems also struggle with medical document processing and clinical usability. We developed KidneyTalk-open, a desktop system integrating three technical components: 1) No-code deployment of state-of-the-art (SOTA) open-source LLMs (such as DeepSeek-r1, Qwen2.5) via local inference engine; 2) Medical document processing pipeline combining context-aware chunking and intelligent filtering; 3) Adaptive Retrieval and Augmentation Pipeline (AddRep) employing agents collaboration for improving the recall rate of medical documents. A graphical interface was designed to enable clinicians to manage medical documents and conduct AI-powered consultations without technical expertise. Experimental validation on 1,455 challenging nephrology exam questions demonstrates AddRep's effectiveness: achieving 29.1% accuracy (+8.1% over baseline) with intelligent knowledge integration, while maintaining robustness through 4.9% rejection rate to suppress hallucinations. Comparative case studies with the mainstream products (AnythingLLM, Chatbox, GPT4ALL) demonstrate KidneyTalk-open's superior performance in real clinical query. KidneyTalk-open represents the first no-code medical LLM system enabling secure documentation-enhanced medical Q&A on desktop. Its designs establishes a new framework for privacy-sensitive clinical AI applications. The system significantly lowers technical barriers while improving evidence traceability, enabling more medical staff or patients to use SOTA open-source LLMs conveniently.

Yongchao Long, Xian Wu, Yingying Zhang et al.

While Retrieval-Augmented Generation (RAG) enables large language models (LLMs) to generate contextually grounded responses, contextual faithfulness remains challenging as LLMs may not consistently trust provided context, leading to hallucinations that undermine reliability. We observe an inverse correlation between response copying degree and context-unfaithful hallucinations on RAGTruth, suggesting that higher copying degrees reduce hallucinations by fostering genuine contextual belief. We propose CopyPasteLLM, obtained through two-stage high-copying response preference training. We design three prompting methods to enhance copying degree, demonstrating that high-copying responses achieve superior contextual faithfulness and hallucination control. These approaches enable a fully automated pipeline that transforms generated responses into high-copying preference data for training CopyPasteLLM. On FaithEval, ConFiQA and PubMedQA, CopyPasteLLM achieves best performance in both counterfactual and original contexts, remarkably with 12.2% to 24.5% accuracy improvements on FaithEval over the best baseline, while requiring only 365 training samples -- 1/50th of baseline data. To elucidate CopyPasteLLM's effectiveness, we propose the Context-Parameter Copying Capturing algorithm. Interestingly, this reveals that CopyPasteLLM recalibrates reliance on internal parametric knowledge rather than external knowledge during generation. All codes are available at https://github.com/longyongchao/CopyPasteLLM

Zhengbo Zhang, Yuxi Zhou, Duo Peng et al.

One-shot controllable video editing (OCVE) is an important yet challenging task, aiming to propagate user edits that are made -- using any image editing tool -- on the first frame of a video to all subsequent frames, while ensuring content consistency between edited frames and source frames. To achieve this, prior methods employ DDIM inversion to transform source frames into latent noise, which is then fed into a pre-trained diffusion model, conditioned on the user-edited first frame, to generate the edited video. However, the DDIM inversion process accumulates errors, which hinder the latent noise from accurately reconstructing the source frames, ultimately compromising content consistency in the generated edited frames. To overcome it, our method eliminates the need for DDIM inversion by performing OCVE through a novel perspective based on visual prompting. Furthermore, inspired by consistency models that can perform multi-step consistency sampling to generate a sequence of content-consistent images, we propose a content consistency sampling (CCS) to ensure content consistency between the generated edited frames and the source frames. Moreover, we introduce a temporal-content consistency sampling (TCS) based on Stein Variational Gradient Descent to ensure temporal consistency across the edited frames. Extensive experiments validate the effectiveness of our approach.

Dongqian Guo, Wencheng Han, Pang Lyu et al.

Cephalometric landmark detection is essential for orthodontic diagnostics and treatment planning. Nevertheless, the scarcity of samples in data collection and the extensive effort required for manual annotation have significantly impeded the availability of diverse datasets. This limitation has restricted the effectiveness of deep learning-based detection methods, particularly those based on large-scale vision models. To address these challenges, we have developed an innovative data generation method capable of producing diverse cephalometric X-ray images along with corresponding annotations without human intervention. To achieve this, our approach initiates by constructing new cephalometric landmark annotations using anatomical priors. Then, we employ a diffusion-based generator to create realistic X-ray images that correspond closely with these annotations. To achieve precise control in producing samples with different attributes, we introduce a novel prompt cephalometric X-ray image dataset. This dataset includes real cephalometric X-ray images and detailed medical text prompts describing the images. By leveraging these detailed prompts, our method improves the generation process to control different styles and attributes. Facilitated by the large, diverse generated data, we introduce large-scale vision detection models into the cephalometric landmark detection task to improve accuracy. Experimental results demonstrate that training with the generated data substantially enhances the performance. Compared to methods without using the generated data, our approach improves the Success Detection Rate (SDR) by 6.5%, attaining a notable 82.2%. All code and data are available at: https://um-lab.github.io/cepha-generation

Gangxiong Zhang, Yongchao Long, Yong Zhang et al.

Accurate mortality risk prediction for intensive care unit (ICU) patients is essential for clinical decision-making. Although large language models (LLMs) show promise in predicting outcomes from structured medical data, their predictions may exhibit demographic biases related to sex, age, and race, limiting their trustworthy use in clinical practice. Existing debiasing methods often reduce predictive performance, making it difficult to jointly optimize fairness and accuracy. In this study, we systematically examine bias in LLM-based ICU mortality prediction and propose a training-free, clinically adaptive prompting framework to simultaneously improve fairness and performance. We first develop a multi-dimensional bias assessment scheme for comprehensive model diagnosis. Building on this analysis, we introduce CAse Prompting (CAP), a novel prompting framework that integrates conventional debiasing prompts with case-based reasoning. CAP guides the model to learn from similar historical misprediction cases and their correct outcomes, enabling correction of biased reasoning patterns. Experiments on the MIMIC-IV dataset show that CAP substantially improves both predictive accuracy and fairness. CAP increases AUROC from 0.806 to 0.873 and AUPRC from 0.497 to 0.694, while reducing sex- and race-related disparities by over 90%. Feature reliance analysis further indicates highly consistent attention patterns across demographic groups, with similarity scores exceeding 0.98. These results demonstrate that LLMs exhibit measurable bias in ICU mortality prediction, and that a carefully designed prompting framework can effectively co-optimize fairness and performance without retraining, offering a transferable paradigm for equitable clinical decision support.

Bin Liu, Qinghao Zhao, Yuxi Zhou et al.

Right heart failure (RHF) is a disease characterized by abnormalities in the structure or function of the right ventricle (RV), which is associated with high morbidity and mortality. Lung disease often causes increased right ventricular load, leading to RHF. Therefore, it is very important to screen out patients with cor pulmonale who develop RHF from people with underlying lung diseases. In this work, we propose a self-supervised representation learning method to early detecting RHF from patients with cor pulmonale, which uses spirogram time series to predict patients with RHF at an early stage. The proposed model is divided into two stages. The first stage is the self-supervised representation learning-based spirogram embedding (SLSE) network training process, where the encoder of the Variational autoencoder (VAE-encoder) learns a robust low-dimensional representation of the spirogram time series from the data-augmented unlabeled data. Second, this low-dimensional representation is fused with demographic information and fed into a CatBoost classifier for the downstream RHF prediction task. Trained and tested on a carefully selected subset of 26,617 individuals from the UK Biobank, our model achieved an AUROC of 0.7501 in detecting RHF, demonstrating strong population-level distinction ability. We further evaluated the model on high-risk clinical subgroups, achieving AUROC values of 0.8194 on a test set of 74 patients with chronic kidney disease (CKD) and 0.8413 on a set of 64 patients with valvular heart disease (VHD). These results highlight the model's potential utility in predicting RHF among clinically elevated-risk populations. In conclusion, this study presents a self-supervised representation learning approach combining spirogram time series and demographic data, demonstrating promising potential for early RHF detection in clinical practice.

Runze Xia, Yupeng Ji, Yuxi Zhou et al.

Query-service relevance prediction in e-commerce search systems faces strict latency requirements that prevent the direct application of Large Language Models (LLMs). To bridge this gap, we propose a two-stage reasoning distillation framework to transfer reasoning capabilities from a powerful teacher LLM to a lightweight, deployment-friendly student model. In the first stage, we address the limitations of general-purpose LLMs by constructing a domain-adapted teacher model. This is achieved through a three-step process: domain-adaptive pre-training to inject platform knowledge, supervised fine-tuning to elicit reasoning skills, and preference optimization with a multi-dimensional reward model to ensure the generation of reliable and preference-aligned reasoning paths. This teacher can then automatically annotate massive query-service pairs from search logs with both relevance labels and reasoning chains. In the second stage, to address the challenges of architectural heterogeneity in standard distillation, we introduce Contrastive Reasoning Self-Distillation (CRSD). By modeling the behavior of the same student model under "standard" and "reasoning-augmented" inputs as a teacher-student relationship, CRSD enables the lightweight model to internalize the teacher's complex decision-making mechanisms without needing the explicit reasoning path at inference. Offline evaluations and online A/B testing in the Meituan search advertising system demonstrate that our framework achieves significant improvements across multiple metrics, validating its effectiveness and practical value.

Jinshuai Gu, Zenghui Lin, Jingying Ma et al.

Cardiotocography (CTG) is a low-cost, non-invasive fetal health assessment technique used globally, especially in underdeveloped countries. However, it is currently mainly used to identify the fetus's current status (e.g., fetal acidosis or hypoxia), and the potential of CTG in predicting future adverse pregnancy outcomes has not been fully explored. We aim to develop an AI-based model that predicts biological age from CTG time series (named CTGage), then calculate the age gap between CTGage and actual age (named CTGage-gap), and use this gap as a new digital biomarker for future adverse pregnancy outcomes. The CTGage model is developed using 61,140 records from 11,385 pregnant women, collected at Peking University People's Hospital between 2018 and 2022. For model training, a structurally designed 1D convolutional neural network is used, incorporating distribution-aligned augmented regression technology. The CTGage-gap is categorized into five groups: < -21 days (underestimation group), -21 to -7 days, -7 to 7 days (normal group), 7 to 21 days, and > 21 days (overestimation group). We further defined the underestimation group and overestimation group together as the high-risk group. We then compare the incidence of adverse outcomes and maternal diseases across these groups. The average absolute error of the CTGage model is 10.91 days. When comparing the overestimation group with the normal group, premature infants incidence is 5.33% vs. 1.42% (p < 0.05) and gestational diabetes mellitus (GDM) incidence is 31.93% vs. 20.86% (p < 0.05). When comparing the underestimation group with the normal group, low birth weight incidence is 0.17% vs. 0.15% (p < 0.05) and anaemia incidence is 37.51% vs. 34.74% (p < 0.05). Artificial intelligence-derived CTGage can predict the future risk of adverse pregnancy outcomes and hold potential as a novel, non-invasive, and easily accessible digital biomarker.

Shanwei Zhang, Deyun Zhang, Yirao Tao et al.

Electrocardiogram (ECG) as an important tool for diagnosing cardiovascular diseases such as arrhythmia. Due to the differences in ECG layouts used by different hospitals, the digitized signals exhibit asynchronous lead time and partial blackout loss, which poses a serious challenge to existing models. To address this challenge, the study introduced PatchECG, a framework for adaptive variable block count missing representation learning based on a masking training strategy, which automatically focuses on key patches with collaborative dependencies between leads, thereby achieving key recognition of arrhythmia in ECGs with different layouts. Experiments were conducted on the PTB-XL dataset and 21388 asynchronous ECG images generated using ECG image kit tool, using the 23 Subclasses as labels. The proposed method demonstrated strong robustness under different layouts, with average Area Under the Receiver Operating Characteristic Curve (AUROC) of 0.835 and remained stable (unchanged with layout changes). In external validation based on 400 real ECG images data from Chaoyang Hospital, the AUROC for atrial fibrillation diagnosis reached 0.778; On 12 x 1 layout ECGs, AUROC reaches 0.893. This result is superior to various classic interpolation and baseline methods, and compared to the current optimal large-scale pre-training model ECGFounder, it has improved by 0.111 and 0.19.

Shuhao Mei, Yongchao Long, Shan Cao et al.

Chronic Obstructive Pulmonary Disease (COPD), a major chronic respiratory disease with persistent airflow limitation, is a leading global cause of disability and mortality. Respiratory spirogram time series, routinely collected during pulmonary function tests (PFTs), play a critical role in the early detection of repsiratory diseases and in monitoring lung function over time. However, most current AI models for COPD diagnosis are limited to outputting classification results without providing a rationale for their diagnostic process, while current Large Language Models (LLMs) cannot understand spirograms yet, which severely limits their clinical trust and adoption. To tackle this challenge, we leverage a cohort of 234,028 individuals from the UK Biobank (UKB) to propose SpiroLLM, the first multimodal large language model that can understand spirogram. The model extracts morphological features from respiratory curves via a SpiroEncoder and aligns them with PFT numerical values in a unified latent space using a SpiroProjector, ultimately empowering a large language model to generate a comprehensive diagnostic report. Experimental results confirm that SpiroLLM achieved a diagnostic AUROC of 0.8980 (95% CI: 0.8820-0.9132). In a robustness test with missing core data, it maintained a 100% valid response rate, far surpassing the 13.4% of a text-only model and showcasing the superiority of its multimodal design. This work demonstrates the substantial potential of deeply fusing physiological signals with large language models, establishing a new paradigm for the next generation of interpretable and reliable clinical decision support tools.

Zhengbo Zhang, Yuxi Zhou, Jia Gong et al.

Knowledge distillation (KD) enhances the performance of a student network by allowing it to learn the knowledge transferred from a teacher network incrementally. Existing methods dynamically adjust the temperature to enable the student network to adapt to the varying learning difficulties at different learning stages of KD. KD is a continuous process, but when adjusting the temperature, these methods consider only the immediate benefits of the operation in the current learning phase and fail to take into account its future returns. To address this issue, we formulate the adjustment of temperature as a sequential decision-making task and propose a method based on reinforcement learning, termed RLKD. Importantly, we design a novel state representation to enable the agent to make more informed action (i.e. instance temperature adjustment). To handle the problem of delayed rewards in our method due to the KD setting, we explore an instance reward calibration approach. In addition,we devise an efficient exploration strategy that enables the agent to learn valuable instance temperature adjustment policy more efficiently. Our framework can serve as a plug-and-play technique to be inserted into various KD methods easily, and we validate its effectiveness on both image classification and object detection tasks. Our project is at https://itkd123.github.io/ITKD.github.io/.

Shuhao Mei, Xin Li, Yuxi Zhou et al.

Chronic Obstructive Pulmonary Disease (COPD) is a chronic lung condition characterized by airflow obstruction. Current diagnostic methods primarily rely on identifying prominent features in spirometry (Volume-Flow time series) to detect COPD, but they are not adept at predicting future COPD risk based on subtle data patterns. In this study, we introduce a novel deep learning-based approach, DeepSpiro, aimed at the early prediction of future COPD risk. DeepSpiro consists of four key components: SpiroSmoother for stabilizing the Volume-Flow curve, SpiroEncoder for capturing volume variability-pattern through key patches of varying lengths, SpiroExplainer for integrating heterogeneous data and explaining predictions through volume attention, and SpiroPredictor for predicting the disease risk of undiagnosed high-risk patients based on key patch concavity, with prediction horizons of 1, 2, 3, 4, 5 years, or even longer. Evaluated on the UK Biobank dataset, DeepSpiro achieved an AUC of 0.8328 for COPD detection and demonstrated strong predictive performance for future COPD risk (p-value < 0.001). In summary, DeepSpiro can effectively predicts the long-term progression of the COPD disease.

Shenda Hong, Yuxi Zhou, Junyuan Shang et al.

Background:The electrocardiogram (ECG) is one of the most commonly used diagnostic tools in medicine and healthcare. Deep learning methods have achieved promising results on predictive healthcare tasks using ECG signals. Objective:This paper presents a systematic review of deep learning methods for ECG data from both modeling and application perspectives. Methods:We extracted papers that applied deep learning (deep neural network) models to ECG data that were published between Jan. 1st of 2010 and Feb. 29th of 2020 from Google Scholar, PubMed, and the DBLP. We then analyzed each article according to three factors: tasks, models, and data. Finally, we discuss open challenges and unsolved problems in this area. Results: The total number of papers extracted was 191. Among these papers, 108 were published after 2019. Different deep learning architectures have been used in various ECG analytics tasks, such as disease detection/classification, annotation/localization, sleep staging, biometric human identification, and denoising. Conclusion: The number of works on deep learning for ECG data has grown explosively in recent years. Such works have achieved accuracy comparable to that of traditional feature-based approaches and ensembles of multiple approaches can achieve even better results. Specifically, we found that a hybrid architecture of a convolutional neural network and recurrent neural network ensemble using expert features yields the best results. However, there are some new challenges and problems related to interpretability, scalability, and efficiency that must be addressed. Furthermore, it is also worth investigating new applications from the perspectives of datasets and methods. Significance: This paper summarizes existing deep learning research using ECG data from multiple perspectives and highlights existing challenges and problems to identify potential future research directions.