arXiv subjects

Yi Qin

Publications and source records attributed to Yi Qin.

At least 19 recordsLinked to original sources

arXiv2026

Drag-induced skin effect in a Bose-Fermi mixture

The non-Hermitian skin effect (NHSE) represents one of the most distinctive phenomena in non-Hermitian physics. Here, we uncover a new drag-induced NHSE mechanism in interacting Bose--Fermi mixtures where only bosons and not fermions experience asymmetric hoppings. %While bosons exhibit intrinsic skin localization due to asymmetric hopping, fermions remain Hermitian in isolation and do not independently support NHSE. We show that strong Bose--Fermi interactions enable fermions to inherit boundary accumulation through correlated bound states. %In the few-body regime, The interplay of interactions, quantum statistics, and non-Hermitian dynamics gives rise to an interaction-induced blockade mechanism, leading to highly asymmetric fermionic transport. We demonstrate that the drag-induced NHSE is dynamically stable and propose a feasible realization in ultracold Bose--Fermi mixtures with Floquet-engineered asymmetric tunneling. Our results establish a general interaction-mediated mechanism for emergent non-Hermitian localization in hybrid quantum matter.

cond-mat.quant-gas↗

arXiv2026

TriALS: Triphasic-Aided Liver Lesion Segmentation Benchmark in Non-Contrast CT

Automated segmentation of liver lesions on non-contrast computed tomography (NCCT) is clinically important but fundamentally challenging, particularly in low-resource settings across Africa and Asia where contrast agents are frequently unavailable. Progress has been limited by the absence of annotated NCCT benchmarks. Here we describe the TriALS challenge for automated liver lesion segmentation under contrast-limited conditions, supported by a multi-centre dataset of 150 cases with four-phase CT acquisitions (600 volumes) from Egyptian and Chinese institutions. Algorithms were evaluated on 70 cases from three institutions, including an independent external cohort. The top-performing method achieved a mean venous-phase Dice of 0.754, consistent with human-level performance, yet dropped to 0.57 on NCCT. On external validation, the leading method outperformed off-the-shelf models by up to 28% in Dice on NCCT. Algorithm performance was most strongly predicted by training data scale and pre-training strategy. A cross-year comparison exposed a persistent perceptual barrier on NCCT that scaling pre-training alone cannot overcome. Data, annotations, and code are available at https://github.com/xmed-lab/TriALS.

cs.CV↗

arXiv2026

Anyon-Induced Criticality and Dynamical Stability in Non-Hermitian Many-Body Systems

We show that anyonic statistics fundamentally reshapes non-Hermitian many-body physics by intrinsically breaking pseudo-Hermiticity, leading to a unique real-complex spectral transition with characteristically dense states in Im$E$. This anyon-induced transition occurs even when bosonic and pseudofermionic counterparts remain entirely real, revealing a form of non-Hermitian criticality driven purely by exchange statistics. The resulting spectrum exhibits enhanced gaps in Im$E$ that dynamically isolate dominant eigenstates, producing anomalously stable short-time quench dynamics for anyons. Our results identify anyonic statistics as an intrinsic mechanism for generating unconventional non-Hermitian critical behavior usually associated with highly non-local systems.

quant-ph↗

arXiv2026

Detecting Misbehaviors of Large Vision-Language Models by Evidential Uncertainty Quantification

%Large vision-language models (LVLMs) have shown substantial advances in multimodal understanding and generation. However, when presented with incompetent or adversarial inputs, they frequently produce unreliable or even harmful content, such as fact hallucinations or dangerous instructions. This misalignment with human expectations, referred to as \emph{misbehaviors} of LVLMs, raises serious concerns for deployment in critical applications. These misbehaviors are found to stem from epistemic uncertainty, specifically either conflicting internal knowledge or the absence of supporting information. However, existing uncertainty quantification methods, which typically capture only overall epistemic uncertainty, have shown limited effectiveness in identifying such issues. To address this gap, we propose Evidential Uncertainty Quantification (EUQ), a fine-grained method that captures both information conflict and ignorance for effective detection of LVLM misbehaviors. In particular, we interpret features from the model output head as either supporting (positive) or opposing (negative) evidence. Leveraging Evidence Theory, we model and aggregate this evidence to quantify internal conflict and knowledge gaps within a single forward pass. %We extensively evaluate our method across four categories of misbehavior, including hallucinations, jailbreaks, adversarial vulnerabilities, and out-of-distribution (OOD) failures, using state-of-the-art LVLMs, and find that EUQ consistently outperforms strong baselines, showing that hallucinations correspond to high internal conflict and OOD failures to high ignorance. Furthermore, layer-wise evidential uncertainty dynamics analysis helps interpret the evolution of internal representations from a new perspective. The source code is available at https://github.com/HT86159/EUQ.

cs.LG↗

arXiv2026

Incentivizing Cardiologist-Like Reasoning in MLLMs for Interpretable Echocardiographic Diagnosis

Echocardiographic diagnosis is vital for cardiac screening yet remains challenging. Existing echocardiography foundation models do not effectively capture the relationships between quantitative measurements and clinical manifestations, whereas medical reasoning multimodal large language models (MLLMs) require costly construction of detailed reasoning paths and remain ineffective at directly incorporating such echocardiographic priors into their reasoning. To address these limitations, we propose a novel approach comprising Cardiac Reasoning Template (CRT) and CardiacMind to enhance MLLM's echocardiographic reasoning by introducing cardiologist-like mindset. Specifically, CRT provides stepwise canonical diagnostic procedures for complex cardiac diseases to streamline reasoning path construction without the need for costly case-by-case verification. To incentivize reasoning MLLM under CRT, we develop CardiacMind, a new reinforcement learning scheme with three novel rewards: Procedural Quantity Reward (PQtR), Procedural Quality Reward (PQlR), and Echocardiographic Semantic Reward (ESR). PQtR promotes detailed reasoning; PQlR promotes integration of evidence across views and modalities, while ESR grounds stepwise descriptions in visual content. Our methods show a 48% improvement in multiview echocardiographic diagnosis for 15 complex cardiac diseases and a 5% improvement on CardiacNet-PAH over prior methods. The user study on our method's reasoning outputs shows 93.33% clinician agreement with cardiologist-like reasoning logic. Our code will be available.

cs.CV↗

arXiv2025

Proactive Reasoning-with-Retrieval Framework for Medical Multimodal Large Language Models

Incentivizing the reasoning ability of Multimodal Large Language Models (MLLMs) is essential for medical applications to transparently analyze medical scans and provide reliable diagnosis. However, existing medical MLLMs rely solely on internal knowledge during reasoning, leading to hallucinated reasoning and factual inaccuracies when encountering cases beyond their training scope. Although recent Agentic Retrieval-Augmented Generation (RAG) methods elicit the medical model's proactive retrieval ability during reasoning, they are confined to unimodal LLMs, neglecting the crucial visual information during reasoning and retrieval. Consequently, we propose the first Multimodal Medical Reasoning-with-Retrieval framework, Med-RwR, which actively retrieves external knowledge by querying observed symptoms or domain-specific medical concepts during reasoning. Specifically, we design a two-stage reinforcement learning strategy with tailored rewards that stimulate the model to leverage both visual diagnostic findings and textual clinical information for effective retrieval. Building on this foundation, we further propose a Confidence-Driven Image Re-retrieval (CDIR) method for test-time scaling when low prediction confidence is detected. Evaluation on various public medical benchmarks demonstrates Med-RwR's significant improvements over baseline models, proving the effectiveness of enhancing reasoning capabilities with external knowledge integration. Furthermore, Med-RwR demonstrates remarkable generalizability to unfamiliar domains, evidenced by 8.8% performance gain on our proposed EchoCardiography Benchmark (ECBench), despite the scarcity of echocardiography data in the training corpus. Our data, model, and codes will be made publicly available at https://github.com/xmed-lab/Med-RwR.

cs.CV↗

arXiv2025

Echo-Path: Pathology-Conditioned Echo Video Generation

Cardiovascular diseases (CVDs) remain the leading cause of mortality globally, and echocardiography is critical for diagnosis of both common and congenital cardiac conditions. However, echocardiographic data for certain pathologies are scarce, hindering the development of robust automated diagnosis models. In this work, we propose Echo-Path, a novel generative framework to produce echocardiogram videos conditioned on specific cardiac pathologies. Echo-Path can synthesize realistic ultrasound video sequences that exhibit targeted abnormalities, focusing here on atrial septal defect (ASD) and pulmonary arterial hypertension (PAH). Our approach introduces a pathology-conditioning mechanism into a state-of-the-art echo video generator, allowing the model to learn and control disease-specific structural and motion patterns in the heart. Quantitative evaluation demonstrates that the synthetic videos achieve low distribution distances, indicating high visual fidelity. Clinically, the generated echoes exhibit plausible pathology markers. Furthermore, classifiers trained on our synthetic data generalize well to real data and, when used to augment real training sets, it improves downstream diagnosis of ASD and PAH by 7\% and 8\% respectively. Code, weights and dataset are available here https://github.com/Marshall-mk/EchoPathv1

cs.CV↗

arXiv2025

Cross-view Generalized Diffusion Model for Sparse-view CT Reconstruction

Sparse-view computed tomography (CT) reduces radiation exposure by subsampling projection views, but conventional reconstruction methods produce severe streak artifacts with undersampled data. While deep-learning-based methods enable single-step artifact suppression, they often produce over-smoothed results under significant sparsity. Though diffusion models improve reconstruction via iterative refinement and generative priors, they require hundreds of sampling steps and struggle with stability in highly sparse regimes. To tackle these concerns, we present the Cross-view Generalized Diffusion Model (CvG-Diff), which reformulates sparse-view CT reconstruction as a generalized diffusion process. Unlike existing diffusion approaches that rely on stochastic Gaussian degradation, CvG-Diff explicitly models image-domain artifacts caused by angular subsampling as a deterministic degradation operator, leveraging correlations across sparse-view CT at different sample rates. To address the inherent artifact propagation and inefficiency of sequential sampling in generalized diffusion model, we introduce two innovations: Error-Propagating Composite Training (EPCT), which facilitates identifying error-prone regions and suppresses propagated artifacts, and Semantic-Prioritized Dual-Phase Sampling (SPDPS), an adaptive strategy that prioritizes semantic correctness before detail refinement. Together, these innovations enable CvG-Diff to achieve high-quality reconstructions with minimal iterations, achieving 38.34 dB PSNR and 0.9518 SSIM for 18-view CT using only \textbf{10} steps on AAPM-LDCT dataset. Extensive experiments demonstrate the superiority of CvG-Diff over state-of-the-art sparse-view CT reconstruction methods. The code is available at https://github.com/xmed-lab/CvG-Diff.

eess.IV↗

arXiv2025

Never compromise with vulnerabilities: a comprehensive survey on AI governance

The rapid advancement of AI has expanded its capabilities across domains, yet introduced critical technical vulnerabilities, such as algorithmic bias and adversarial sensitivity, that pose significant societal risks, including misinformation, inequity, security breaches, physical harm, and eroded public trust. These challenges highlight the urgent need for robust AI governance. We propose a comprehensive framework integrating technical and societal dimensions, structured around three interconnected pillars: Intrinsic Security (system reliability), Derivative Security (real-world harm mitigation), and Social Ethics (value alignment and accountability). Uniquely, our approach unifies technical methods, emerging evaluation benchmarks, and policy insights to promote transparency, accountability, and trust in AI systems. Through a systematic review of over 300 studies, we identify three core challenges: (1) the generalization gap, where defenses fail against evolving threats; (2) inadequate evaluation protocols that overlook real-world risks; and (3) fragmented regulations leading to inconsistent oversight. These shortcomings stem from treating governance as an afterthought, rather than a foundational design principle, resulting in reactive, siloed efforts that fail to address the interdependence of technical integrity and societal trust. To overcome this, we present an integrated research agenda that bridges technical rigor with social responsibility. Our framework offers actionable guidance for researchers, engineers, and policymakers to develop AI systems that are not only robust and secure but also ethically aligned and publicly trustworthy. The accompanying repository is available at https://github.com/Tele-EVOL/AI-Governance.

cs.CR↗

arXiv2025

SAM Encoder Breach by Adversarial Simplicial Complex Triggers Downstream Model Failures

While the Segment Anything Model (SAM) transforms interactive segmentation with zero-shot abilities, its inherent vulnerabilities present a single-point risk, potentially leading to the failure of numerous downstream applications. Proactively evaluating these transferable vulnerabilities is thus imperative. Prior adversarial attacks on SAM often present limited transferability due to insufficient exploration of common weakness across domains. To address this, we propose Vertex-Refining Simplicial Complex Attack (VeSCA), a novel method that leverages only the encoder of SAM for generating transferable adversarial examples. Specifically, it achieves this by explicitly characterizing the shared vulnerable regions between SAM and downstream models through a parametric simplicial complex. Our goal is to identify such complexes within adversarially potent regions by iterative vertex-wise refinement. A lightweight domain re-adaptation strategy is introduced to bridge domain divergence using minimal reference data during the initialization of simplicial complex. Ultimately, VeSCA generates consistently transferable adversarial examples through random simplicial complex sampling. Extensive experiments demonstrate that VeSCA achieves performance improved by 12.7% compared to state-of-the-art methods across three downstream model categories across five domain-specific datasets. Our findings further highlight the downstream model risks posed by SAM's vulnerabilities and emphasize the urgency of developing more robust foundation models.

cs.CV↗

arXiv2025

Many-body critical non-Hermitian skin effect

Criticality in non-Hermitian systems unveils unique phase transitions and scaling behaviors beyond Hermitian paradigms, offering new insights into the interplay between gain/loss, non-reciprocity, and complex energy spectra. In this paper, we uncover a new class of many-body critical non-Hermitian skin effect (CSE) originating from the interplay between multiple non-Hermitian pumping channels and Hubbard interactions. In particular, criticality in the real-to-complex transitions can selectively emerge within the subspace of bound states or scattering states, as well as their interacting admixtures. These mechanisms possess no single-particle analog and can be diagnosed through a specially defined correlation function. As more particles are involved, higher-order CSEs naturally arise, with greatly enhanced effective coupling strengths and hence greater experimental accessibility. Our results reveal an enriched landscape of non-Hermitian critical phenomena in interacting many-body systems, and pave the way for investigating unconventional non-Hermitian criticality in the context of various interaction-induced particle clustering configurations.

quant-ph↗

arXiv2025

Multi-Modal Explainable Medical AI Assistant for Trustworthy Human-AI Collaboration

Generalist Medical AI (GMAI) systems have demonstrated expert-level performance in biomedical perception tasks, yet their clinical utility remains limited by inadequate multi-modal explainability and suboptimal prognostic capabilities. Here, we present XMedGPT, a clinician-centric, multi-modal AI assistant that integrates textual and visual interpretability to support transparent and trustworthy medical decision-making. XMedGPT not only produces accurate diagnostic and descriptive outputs, but also grounds referenced anatomical sites within medical images, bridging critical gaps in interpretability and enhancing clinician usability. To support real-world deployment, we introduce a reliability indexing mechanism that quantifies uncertainty through consistency-based assessment via interactive question-answering. We validate XMedGPT across four pillars: multi-modal interpretability, uncertainty quantification, and prognostic modeling, and rigorous benchmarking. The model achieves an IoU of 0.703 across 141 anatomical regions, and a Kendall's tau-b of 0.479, demonstrating strong alignment between visual rationales and clinical outcomes. For uncertainty estimation, it attains an AUC of 0.862 on visual question answering and 0.764 on radiology report generation. In survival and recurrence prediction for lung and glioma cancers, it surpasses prior leading models by 26.9%, and outperforms GPT-4o by 25.0%. Rigorous benchmarking across 347 datasets covers 40 imaging modalities and external validation spans 4 anatomical systems confirming exceptional generalizability, with performance gains surpassing existing GMAI by 20.7% for in-domain evaluation and 16.7% on 11,530 in-house data evaluation. Together, XMedGPT represents a significant leap forward in clinician-centric AI integration, offering trustworthy and scalable support for diverse healthcare applications.

cs.CV↗

arXiv2025

Concept-Based Unsupervised Domain Adaptation

Concept Bottleneck Models (CBMs) enhance interpretability by explaining predictions through human-understandable concepts but typically assume that training and test data share the same distribution. This assumption often fails under domain shifts, leading to degraded performance and poor generalization. To address these limitations and improve the robustness of CBMs, we propose the Concept-based Unsupervised Domain Adaptation (CUDA) framework. CUDA is designed to: (1) align concept representations across domains using adversarial training, (2) introduce a relaxation threshold to allow minor domain-specific differences in concept distributions, thereby preventing performance drop due to over-constraints of these distributions, (3) infer concepts directly in the target domain without requiring labeled concept data, enabling CBMs to adapt to diverse domains, and (4) integrate concept learning into conventional domain adaptation (DA) with theoretical guarantees, improving interpretability and establishing new benchmarks for DA. Experiments demonstrate that our approach significantly outperforms the state-of-the-art CBM and DA methods on real-world datasets.

cs.LG↗

arXiv2025

Synergistic Vision-Language Reinforcement Enables Scalable On-Demand Analysis across Diverse Clinical Tasks

Accurate delineation of tumors and surrounding organs-at-risk is essential for radiotherapy, surgery and treatment response assessment, yet remains time-consuming and expertise-intensive. Existing artificial intelligence systems often require manual spatial prompts or task-specific retraining, while generic class labels provide limited semantic grounding for heterogeneous disease targets. Here we present SyRe, a promptable segmentation foundation model based on Synergistic vision-language Reinforcement. SyRe strengthens bidirectional interaction between visual and linguistic representations to improve semantically grounded spatial understanding. To support large-scale training, we introduce the Color Region Description strategy and construct SyReData, comprising 20 million image-mask-description triplets across 9 modalities and 229 segmentation tasks. Training with diversified prompt forms further enables open-ended prompting, invalid-prompt rejection and flexible switching between single- and multi-target analysis. SyRe achieves accurate text-prompted segmentation across diverse clinical scenarios, with particularly strong performance on disease-related targets. Across 28 unseen external datasets, including 20 cancer types and multinational in-house cohorts, SyRe generalizes robustly under real-world distribution shifts. SyRe-generated masks also preserve clinically relevant quantitative information in pathology and yield radiomics features that stratify survival and improve prognostic modeling across five retrospective CT and MRI tumor cohorts. Finally, clinician-in-the-loop refinement enables efficient case-level correction when greater precision is required. These results establish SyRe as a generalizable foundation for scalable quantitative oncology and clinician-guided segmentation refinement.

cs.CV↗

arXiv2025

Feature Selection Based on Reinforcement Learning and Hazard State Classification for Magnetic Adhesion Wall-Climbing Robots

Magnetic adhesion tracked wall-climbing robots face potential risks of overturning during high-altitude operations, making their stability crucial for ensuring safety. This study presents a dynamic feature selection method based on Proximal Policy Optimization (PPO) reinforcement learning, combined with typical machine learning models, aimed at improving the classification accuracy of hazardous states under complex operating conditions. Firstly, this work innovatively employs a fiber rod-based MEMS attitude sensor to collect vibration data from the robot and extract high-dimensional feature vectors in both time and frequency domains. Then, a reinforcement learning model is used to dynamically select the optimal feature subset, reducing feature redundancy and enhancing classification accuracy. Finally, a CNN-LSTM deep learning model is employed for classification and recognition. Experimental results demonstrate that the proposed method significantly improves the robot's ability to assess hazardous states across various operational scenarios, providing reliable technical support for robotic safety monitoring.

cs.RO↗

arXiv2024

Improved ICNN-LSTM Model Classification Based on Attitude Sensor Data for Hazardous State Assessment of Magnetic Adhesion Climbing Wall Robots

Magnetic adhesion tracked climbing robots are widely utilized in high-altitude inspection, welding, and cleaning tasks due to their ability to perform various operations against gravity on vertical or inclined walls. However, during operation, the robot may experience overturning torque caused by its own weight and load, which can lead to the detachment of magnetic plates and subsequently pose safety risks. This paper proposes an improved ICNN-LSTM network classification method based on Micro-Electro-Mechanical Systems (MEMS) attitude sensor data for real-time monitoring and assessment of hazardous states in magnetic adhesion tracked climbing robots. Firstly, a data acquisition strategy for attitude sensors capable of capturing minute vibrations is designed. Secondly, a feature extraction and classification model combining an Improved Convolutional Neural Network (ICNN) with a Long Short-Term Memory (LSTM) network is proposed. Experimental validation demonstrates that the proposed minute vibration sensing method achieves significant results, and the proposed classification model consistently exhibits high accuracy compared to other models. The research findings provide effective technical support for the safe operation of climbing robots

cs.RO↗

arXiv2024

FoAM: Foresight-Augmented Multi-Task Imitation Policy for Robotic Manipulation

Multi-task imitation learning (MTIL) has shown significant potential in robotic manipulation by enabling agents to perform various tasks using a single policy. This simplifies the policy deployment and enhances the agent's adaptability across different scenarios. However, key challenges remain, such as maintaining action reliability (e.g., avoiding abnormal action sequences that deviate from nominal task trajectories) and generalizing to unseen tasks with a few expert demonstrations. To address these challenges, we introduce the Foresight-Augmented Manipulation Policy (FoAM), a novel MTIL policy that pioneers the use of multi-modal goal condition as input and introduces a foresight augmentation in addition to the general action reconstruction. FoAM enables the agent to reason about the visual consequences (states) of its actions and learn more expressive embedding that captures nuanced task variations. Extensive experiments on over 100 tasks in simulation and real-world settings demonstrate that FoAM significantly enhances MTIL policy performance, outperforming state-of-the-art baselines by up to 41% in success rate. Meanwhile, we released our simulation suites, including a total of 10 scenarios and over 80 challenging tasks designed for manipulation policy training and evaluation. See the project homepage projFoAM.github.io for project details.

cs.RO↗

arXiv2024

A high-order, high-efficiency adaptive time filter algorithm for shale reservoir model based on coupled fluid flow with porous media flow

In this paper, a third-order time adaptive algorithm with less computation, low complexity is provided for shale reservoir model based on coupled fluid flow with porous media flow. The algorithm combines the three-step linear time filters method for simple post-processing and the second-order backward differential formula (BDF2), is third-order accurate and provides, at no extra computational complexity. At the same time, the time filter method can also be used to damp non-physical oscillations inherent in the BDF2 method, ensuring stability. We proves the variable time stepsize second-order backward differential formula plus time filter (BDF2-TF) algorithm's stability and the convergence properties of the fluid velocity u and hydraulic head $ϕ$ in the $L^2$ norm with an order of $O(k_{n+1}^3 + h^3)$. In the experiments, the adaptive algorithm automatically adjusts the time step in response to the varying characteristics of different models, ensuring that errors are maintained within acceptable limits. This algorithm addresses the issue that high-order algorithms may select inappropriate time steps, resulting in instability or reduced precision of the numerical solution, thereby enhancing calculation accuracy and efficiency. We perform three-dimensional numerical experiments to verify the BDF2-TF algorithm's effectiveness, stability, and third-order convergence. Simultaneously, a simplified model is employed to simulate the process of shale oil extraction from reservoirs, further demonstrating the algorithm's practical applicability.

math.NA↗