Searcharxiv⌕ Search

arXiv subjects

Xian Yang

Publications and source records attributed to Xian Yang.

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

Knowledge-Augmented Multimodal Clinical Rationale Generation for Disease Diagnosis with Small Language Models

Interpretation is critical for disease diagnosis, but existing models struggle to balance predictive accuracy with human-understandable rationales. While large language models (LLMs) offer strong reasoning abilities, their clinical use is limited by high computational costs and restricted multimodal reasoning ability. Small language models (SLMs) are efficient but lack advanced reasoning for integrating multimodal medical data. In addition, both LLMs and SLMs lack domain knowledge for trustworthy reasoning. Therefore, we propose ClinRaGen, enhancing SLMs by leveraging LLM-derived reasoning ability via rationale distillation and domain knowledge injection for trustworthy multimodal rationale generation. Key innovations include a sequential rationale distillation framework that equips SLMs with LLM-comparable multimodal reasoning abilities, and a knowledge-augmented attention mechanism that jointly unifies multimodal representation from time series and textual data in the same encoding space, enabling it to be naturally interpreted by SLMs while incorporating domain knowledge for reliable rationale generation. Experiments on real-world medical datasets show that ClinRaGen achieves state-of-the-art performance in disease diagnosis and rationale generation, demonstrating the effectiveness of combining LLM-driven reasoning with knowledge augmentation for improved interpretability.

cs.CL↗
arXiv2025

ProMedTS: A Self-Supervised, Prompt-Guided Multimodal Approach for Integrating Medical Text and Time Series

Large language models (LLMs) have shown remarkable performance in vision-language tasks, but their application in the medical field remains underexplored, particularly for integrating structured time series data with unstructured clinical notes. In clinical practice, dynamic time series data, such as lab test results, capture critical temporal patterns, while clinical notes provide rich semantic context. Merging these modalities is challenging due to the inherent differences between continuous signals and discrete text. To bridge this gap, we introduce ProMedTS, a novel self-supervised multimodal framework that employs prompt-guided learning to unify these heterogeneous data types. Our approach leverages lightweight anomaly detection to generate anomaly captions that serve as prompts, guiding the encoding of raw time series data into informative prompt embeddings. These prompt embeddings are aligned with textual representations in a shared latent space, preserving fine-grained temporal nuances alongside semantic insights. Furthermore, our framework incorporates tailored self-supervised objectives to enhance both intra- and inter-modal alignment. We evaluate ProMedTS on disease diagnosis tasks using real-world datasets, and the results demonstrate that our method consistently outperforms state-of-the-art approaches.

cs.CL↗
arXiv2025

Progressive Local Alignment for Medical Multimodal Pre-training

Local alignment between medical images and text is essential for accurate diagnosis, though it remains challenging due to the absence of natural local pairings and the limitations of rigid region recognition methods. Traditional approaches rely on hard boundaries, which introduce uncertainty, whereas medical imaging demands flexible soft region recognition to handle irregular structures. To overcome these challenges, we propose the Progressive Local Alignment Network (PLAN), which designs a novel contrastive learning-based approach for local alignment to establish meaningful word-pixel relationships and introduces a progressive learning strategy to iteratively refine these relationships, enhancing alignment precision and robustness. By combining these techniques, PLAN effectively improves soft region recognition while suppressing noise interference. Extensive experiments on multiple medical datasets demonstrate that PLAN surpasses state-of-the-art methods in phrase grounding, image-text retrieval, object detection, and zero-shot classification, setting a new benchmark for medical image-text alignment.

cs.CV↗
arXiv2025

Endo-CLIP: Progressive Self-Supervised Pre-training on Raw Colonoscopy Records

Pre-training on image-text colonoscopy records offers substantial potential for improving endoscopic image analysis, but faces challenges including non-informative background images, complex medical terminology, and ambiguous multi-lesion descriptions. We introduce Endo-CLIP, a novel self-supervised framework that enhances Contrastive Language-Image Pre-training (CLIP) for this domain. Endo-CLIP's three-stage framework--cleansing, attunement, and unification--addresses these challenges by (1) removing background frames, (2) leveraging large language models to extract clinical attributes for fine-grained contrastive learning, and (3) employing patient-level cross-attention to resolve multi-polyp ambiguities. Extensive experiments demonstrate that Endo-CLIP significantly outperforms state-of-the-art pre-training methods in zero-shot and few-shot polyp detection and classification, paving the way for more accurate and clinically relevant endoscopic analysis.

cs.CV↗
arXiv2025

Broken symmetries associated with a Kagome chiral charge order

Chirality or handedness manifests in all fields of science, ranging from cell biology, molecular interaction, and catalysis to different branches of physics. In condensed matter physics, chirality is intrinsic to enigmatic quantum phases, such as chiral charge density waves and chiral superconductivity. Here, the underlying chiral response is subtle and leads to broken symmetries in the ground state. Detection of subtle broken symmetries is the key to understand these quantum states but they are extremely challenging to expose leading to debate and controversy. Here, using second-order optical response, we uncover the broken symmetries of a chiral charge density wave in the Kagome lattice KV3Sb5, revealing the relevant broken symmetries of its charge order. KV3Sb5 undergoes a phase transition to a charge-ordered state at low temperatures. Our polarization-dependent mid-infrared photocurrent microscopy reveals an intrinsic, longitudinal helicity-dependent photocurrent associated with the charge order. Our measurements, supported by our theoretical analysis, provide direct evidence for broken inversion and mirror symmetries at the charge order transition, indicating a chiral charge ordered state. On the other hand, we do not observe a circular photogalvanic effect along the direction perpendicular to that of the incident light, imposing stringent constraints on the rotational and point group symmetries of the charge order. Our study not only visualizes the chiral nature of the Kagome charge order revealing its broken symmetries, but also highlights the nonlinear photogalvanic effect as a sensitive probe for detecting subtle symmetry breakings.

cond-mat.str-el↗
arXiv2025

Field induced density wave in a kagome superconductor

On the kagome lattice, electrons benefit from the simultaneous presence of band topology, flat electronic bands, and van Hove singularities, forming competing or cooperating orders. Understanding the interrelation between these distinct order parameters remains a significant challenge, leaving much of the associated physics unexplored. In the kagome superconductor KV3Sb5, which exhibits a charge density wave (CDW) state below T = 78 K, we uncover an unpredicted field-induced phase transition below 6 K. The observed transition is marked by a hysteretic anomaly in the resistivity, nonlinear electrical transport, and a change in the symmetry of the electronic response as probed via the angular dependence of the magnetoresistivity. These observations surprisingly suggest the emergence of an unanticipated broken symmetry state coexisting with the original CDW. To understand this experimental observation, we developed a theoretical minimal model for the normal state inside the high-temperature parent CDW phase where an incommensurate CDW order emerges as an instability sub-leading to superconductivity. The incommensurate CDW emerges when superconducting fluctuations become fully suppressed by large magnetic fields. Our results suggest that, in kagome superconductors, quantum states can either coexist or are nearly degenerate in energy, indicating that these are rich platforms to expose new correlated phenomena.

cond-mat.str-el↗
arXiv2025

MinMo: A Multimodal Large Language Model for Seamless Voice Interaction

Recent advancements in large language models (LLMs) and multimodal speech-text models have laid the groundwork for seamless voice interactions, enabling real-time, natural, and human-like conversations. Previous models for voice interactions are categorized as native and aligned. Native models integrate speech and text processing in one framework but struggle with issues like differing sequence lengths and insufficient pre-training. Aligned models maintain text LLM capabilities but are often limited by small datasets and a narrow focus on speech tasks. In this work, we introduce MinMo, a Multimodal Large Language Model with approximately 8B parameters for seamless voice interaction. We address the main limitations of prior aligned multimodal models. We train MinMo through multiple stages of speech-to-text alignment, text-to-speech alignment, speech-to-speech alignment, and duplex interaction alignment, on 1.4 million hours of diverse speech data and a broad range of speech tasks. After the multi-stage training, MinMo achieves state-of-the-art performance across various benchmarks for voice comprehension and generation while maintaining the capabilities of text LLMs, and also facilitates full-duplex conversation, that is, simultaneous two-way communication between the user and the system. Moreover, we propose a novel and simple voice decoder that outperforms prior models in voice generation. The enhanced instruction-following capabilities of MinMo supports controlling speech generation based on user instructions, with various nuances including emotions, dialects, and speaking rates, and mimicking specific voices. For MinMo, the speech-to-text latency is approximately 100ms, full-duplex latency is approximately 600ms in theory and 800ms in practice. The MinMo project web page is https://funaudiollm.github.io/minmo, and the code and models will be released soon.

cs.CL↗
arXiv2024

Unconventional gapping behavior in a kagome superconductor

Determining the types of superconducting order in quantum materials is a challenge, especially when multiple degrees of freedom, such as bands or orbitals, contribute to the fermiology and when superconductivity competes, intertwines, or coexists with other symmetry-breaking orders. Here, we study the Kagome-lattice superconductor CsV3Sb5, in which multiband superconductivity coexists with a charge order that substantially reduces the compound's space group symmetries. Through a combination of thermodynamic as well as electrical and thermal transport measurements, we uncover two superconducting regimes with distinct transport and thermodynamic characteristics, while finding no evidence for a phase transition separating them. Thermodynamic measurements reveal substantial quasiparticle weight in a high-temperature regime. At lower temperatures, this weight is removed via the formation of a second gap. The two regimes are sharply distinguished by a pronounced enhancement of the upper critical field at low temperatures and by a switch in the anisotropy of the longitudinal thermal conductivity as a function of in-plane magnetic field orientation. We argue that the band with a gap opening at lower temperatures continues to host low-energy quasiparticles, possibly due to a nodal structure of the gap. Taken together, our results present evidence for band-selective superconductivity with remarkable decoupling of the (two) superconducting gaps. The commonly employed multiband scenario, whereby superconductivity emerges in a primary band and is then induced in other bands appears to fail in this unconventional kagome superconductor. Instead, band-selective superconducting pairing is a paradigm that seems to unify seemingly contradicting results in this intensely studied family of materials and beyond.

cond-mat.str-el↗
arXiv2024

EndoFinder: Online Image Retrieval for Explainable Colorectal Polyp Diagnosis

Determining the necessity of resecting malignant polyps during colonoscopy screen is crucial for patient outcomes, yet challenging due to the time-consuming and costly nature of histopathology examination. While deep learning-based classification models have shown promise in achieving optical biopsy with endoscopic images, they often suffer from a lack of explainability. To overcome this limitation, we introduce EndoFinder, a content-based image retrieval framework to find the 'digital twin' polyp in the reference database given a newly detected polyp. The clinical semantics of the new polyp can be inferred referring to the matched ones. EndoFinder pioneers a polyp-aware image encoder that is pre-trained on a large polyp dataset in a self-supervised way, merging masked image modeling with contrastive learning. This results in a generic embedding space ready for different downstream clinical tasks based on image retrieval. We validate the framework on polyp re-identification and optical biopsy tasks, with extensive experiments demonstrating that EndoFinder not only achieves explainable diagnostics but also matches the performance of supervised classification models. EndoFinder's reliance on image retrieval has the potential to support diverse downstream decision-making tasks during real-time colonoscopy procedures.

cs.CV↗
arXiv2024

A Two-Stage Dual-Path Framework for Text Tampering Detection and Recognition

Document tamper detection has always been an important aspect of tamper detection. Before the advent of deep learning, document tamper detection was difficult. We have made some explorations in the field of text tamper detection based on deep learning. Our Ps tamper detection method includes three steps: feature assistance, audit point positioning, and tamper recognition. It involves hierarchical filtering and graded output (tampered/suspected tampered/untampered). By combining artificial tamper data features, we simulate and augment data samples in various scenarios (cropping with noise addition/replacement, single character/space replacement, smearing/splicing, brightness/contrast adjustment, etc.). The auxiliary features include exif/binary stream keyword retrieval/noise, which are used for branch detection based on the results. Audit point positioning uses detection frameworks and controls thresholds for high and low density detection. Tamper recognition employs a dual-path dual-stream recognition network, with RGB and ELA stream feature extraction. After dimensionality reduction through self-correlation percentile pooling, the fused output is processed through vlad, yielding an accuracy of 0.804, recall of 0.659, and precision of 0.913.

cs.CV↗
arXiv2023

Intelligent Parsing: An Automated Parsing Framework for Extracting Design Semantics from E-commerce Creatives

In the industrial e-commerce landscape, creative designs such as banners and posters are ubiquitous. Extracting structured semantic information from creative e-commerce design materials (manuscripts crafted by designers) to obtain design semantics represents a core challenge in the realm of intelligent design. In this paper, we propose a comprehensive automated framework for intelligently parsing creative materials. This framework comprises material recognition, preprocess, smartname, and label layers. The material recognition layer consolidates various detection and recognition interfaces, covering business aspects including detection of auxiliary areas within creative materials and layer-level detection, alongside label identification. Algorithmically, it encompasses a variety of coarse-to-fine methods such as Cascade RCNN, GFL, and other models. The preprocess layer involves filtering creative layers and grading creative materials. The smartname layer achieves intelligent naming for creative materials, while the label layer covers multi-level tagging for creative materials, enabling tagging at different hierarchical levels. Intelligent parsing constitutes a complete parsing framework that significantly aids downstream processes such as intelligent creation, creative optimization, and material library construction. Within the practical business applications at Suning, it markedly enhances the exposure, circulation, and click-through rates of creative materials, expediting the closed-loop production of creative materials and yielding substantial benefits.

cs.CV↗
arXiv2023

Two-stage dynamic creative optimization under sparse ambiguous samples for e-commerce advertising

Ad creative is one of the main mediums for e-commerce advertising. In our approach we decouple this dynamic creative optimization into two stages, a cascaded structure that can trade off between effectiveness and efficiency. In the first stage, we train an automatic creative optimization architecture based on autoco to simulate complex interactions between creative elements. Although we obtained the ranking of different creatives under a sku, because we bucketed and merged historical data according to periods, this confuses the ctr diversity of the same ad creatives on different days and weakens the ability to separate ambiguous samples. Therefore, we propose a transformer-based rerank model. With the help of the rank model, we propose a distillation method to learn the relative order of ideas and extract the ranking knowledge to guide the rerank learning. The creative order soft labels under each sku are generated by the rank model to alleviate the dilemma that a large number of under-represented creatives cannot obtain real labels. Through the knowledge diffusion of rerank, the ambiguous samples are associated with the positive and negative samples. Cascade rerank and autoco to output the estimated value of the synthetic ad image. In the second stage, we designed a bandit model, and the bandit selected one of the output ad of the first stage for timely delivery. Experimental results show that our method can outperform competing baselines in terms of sctr. Online A/B testing shows that our method improves ctr by 10% compared to the baseline.

cs.MM↗
arXiv2023

Smartbanner: Intelligent banner design framework that strikes a balance between creative freedom and design rules

Companies use banners extensively to promote their products, and the intelligent automatic synthesis of banners is a challenging event. Under the premise of inputting only a small amount of information such as product, text and size, it can synthesize styles with high freedom and richness, but at the same time, it must satisfy the design specifications of advertisers for advertising and scenes. We propose an intelligent banner design framework that strikes a balance between creative freedom and design rules, called smartbanner. Smartbanner consists of planner, actuator, adjuster and generator. The banner is synthesized through the combined framework, which fully liberates the designer and reduces the threshold and cost of design. It increases the click-through rate by 30%, improves the human efficiency of designers by 500% under the condition of ensuring the quality of creation, and synthesizes hundreds of millions of pictures in batches throughout the year.

cs.HC↗
arXiv2023

Semi-classical states for fractional Choquard equations with decaying potentials

This paper deals with the following fractional Choquard equation $$\varepsilon^{2s}(-Δ)^su +Vu=\varepsilon^{-α}(I_α*|u|^p)|u|^{p-2}u\ \ \ \mathrm{in}\ \mathbb{R}^N,$$ where $\varepsilon>0$ is a small parameter, $(-Δ)^s$ is the fractional Laplacian, $N>2s$, $s\in(0,1)$, $α\in\big((N-4s)_{+}, N\big)$, $p\in[2, \frac{N+α}{N-2s})$, $I_α$ is a Riesz potential, $V\in C\big(\mathbb{R}^N, [0, +\infty)\big)$ is an electric potential. Under some assumptions on the decay rate of $V$ and the corresponding range of $p$, we prove that the problem has a family of solutions $\{u_\varepsilon\}$ concentrating at a local minimum of $V$ as $\varepsilon\to 0$. Since the potential $V$ decays at infinity, we need to employ a type of penalized argument and implement delicate analysis on the both nonlocal terms to establish regularity, positivity and asymptotic behaviour of $u_\varepsilon$, which is totally different from the local case. As a contrast, we also develop some nonexistence results, which imply that the assumptions on $V$ and $p$ for the existence of $u_\varepsilon$ are almost optimal. To prove our main results, a general strong maximum principle and comparison function for the weak solutions of fractional Laplacian equations are established. The main methods in this paper are variational methods, penalized technique and some comparison principle developed in this paper.

math.AP↗
arXiv2023

Existence and decays of solutions for fractional Schrödinger equations with decaying potentials

We revisit the following fractional Schrödinger equation \begin{align}\label{1a} \varepsilon^{2s}(-Δ)^su +Vu=u^{p-1},\,\,\,u>0,\ \ \ \mathrm{in}\ \R^N, \end{align} where $\varepsilon>0$ is a small parameter, $(-Δ)^s$ denotes the fractional Laplacian, $s\in(0,1)$, $p\in (2, 2_s^*)$, $2_s^*=\frac {2N}{N-2s}$, $N>2s$, $V\in C\big(\R^N, [0, +\infty)\big)$ is a potential. Under various decay assumptions on $V$, we introduce a uniform penalization argument combined with a comparison principle and iteration process to detect an explicit threshold value $p_*$, such that the above problem admits positive concentration solutions if $p\in (p_*, \,2_s^*)$, while it has no positive weak solutions for $p\in (2,\,p_*)$ if $p_*>2$, where the threshold $p_*\in [2, 2^*_s)$ can be characterized explicitly by \begin{equation*}\label{qdj111} p_*=\left\{\begin{array}{l} 2+\frac {2s}{N-2s} \ \ \ \text { if } \lim\limits_{|x| \to \infty} (1+|x|^{2s})V(x)=0,\vspace{1mm} 2+\frac ω{N+2s-ω} \text { if } 0<\inf (1+|x|^ω)V(x)\le \sup (1+|x|^ω)V(x)< \infty \text { for some } ω\in [0, 2s],\vspace{1mm} 2 \ \ \ \ \ \ \ \ \ \ \ \ \ \ \text { if } \inf V(x)\log(e+|x|^2)>0. \end{array}\right. \end{equation*} Moreover, corresponding to the various decay assumptions of $V(x)$, we obtain the decay properties of the solutions at infinity.

math.AP↗
arXiv2022

Imaging real-space flat band localization in kagome magnet FeSn

Kagome lattices host flat bands due to their frustrated lattice geometry, which leads to destructive quantum interference of electron wave functions. Here, we report imaging of the kagome flat band localization in real-space using scanning tunneling microscopy. We identify both the Fe3Sn kagome lattice layer and the Sn2 honeycomb layer with atomic resolution in kagome antiferromagnet FeSn. On the Fe3Sn lattice, at the flat band energy determined by the angle resolved photoemission spectroscopy, tunneling spectroscopy detects an unusual state localized uniquely at the Fe kagome lattice network. We further show that the vectorial in-plane magnetic field manipulates the spatial anisotropy of the localization state within each kagome unit cell. Our results are consistent with the real-space flat band localization in the magnetic kagome lattice. We further discuss the magnetic tuning of flat band localization under the spin-orbit coupled magnetic kagome lattice model.

cond-mat.str-el↗
arXiv2022

Signatures of a topological Weyl loop in Co$_3$Sn$_2$S$_2$

The search for novel topological phases of matter in quantum magnets has emerged as a frontier of condensed matter physics. Here we use state-of-the-art angle-resolved photoemission spectroscopy (ARPES) to investigate single crystals of Co$_3$Sn$_2$S$_2$ in its ferromagnetic phase. We report for the first time signatures of a topological Weyl loop. From fundamental symmetry considerations, this magnetic Weyl loop is expected to be gapless if spin-orbit coupling (SOC) is strictly zero but gapped, with possible Weyl points, under finite SOC. We point out that high-resolution ARPES results to date cannot unambiguously resolve the SOC gap anywhere along the Weyl loop, leaving open the possibility that Co$_3$Sn$_2$S$_2$ hosts zero Weyl points or some non-zero number of Weyl points. On the surface of our samples, we further observe a possible Fermi arc, but we are unable to clearly verify its topological nature using the established counting criteria. As a result, we argue that from the point of view of photoemission spectroscopy the presence of Weyl points and Fermi arcs in Co$_3$Sn$_2$S$_2$ remains ambiguous. Our results have implications for ongoing investigations of Co$_3$Sn$_2$S$_2$ and other topological magnets.

cond-mat.str-el↗
arXiv2022

Label-dependent and event-guided interpretable disease risk prediction using EHRs

Electronic health records (EHRs) contain patients' heterogeneous data that are collected from medical providers involved in the patient's care, including medical notes, clinical events, laboratory test results, symptoms, and diagnoses. In the field of modern healthcare, predicting whether patients would experience any risks based on their EHRs has emerged as a promising research area, in which artificial intelligence (AI) plays a key role. To make AI models practically applicable, it is required that the prediction results should be both accurate and interpretable. To achieve this goal, this paper proposed a label-dependent and event-guided risk prediction model (LERP) to predict the presence of multiple disease risks by mainly extracting information from unstructured medical notes. Our model is featured in the following aspects. First, we adopt a label-dependent mechanism that gives greater attention to words from medical notes that are semantically similar to the names of risk labels. Secondly, as the clinical events (e.g., treatments and drugs) can also indicate the health status of patients, our model utilizes the information from events and uses them to generate an event-guided representation of medical notes. Thirdly, both label-dependent and event-guided representations are integrated to make a robust prediction, in which the interpretability is enabled by the attention weights over words from medical notes. To demonstrate the applicability of the proposed method, we apply it to the MIMIC-III dataset, which contains real-world EHRs collected from hospitals. Our method is evaluated in both quantitative and qualitative ways.

cs.AI↗