arXiv subjects

Xiaochen Zheng

Publications and source records attributed to Xiaochen Zheng.

At least 19 recordsLinked to original sources

arXiv2026

The complex kinematics of the young stars orbiting the supermassive black hole in the Galactic center can be explained by the presence of an intermediate mass companion of Sgr A$^\star$

The sub-parsec proximity around the Sgr A$^\star$ supermassive black hole (SMBH) in the center of the Milky Way contains an inner cluster of eccentric S-stars with randomly oriented orbits, a midway-disk of clockwise-rotating stars (CWSs), and a surrounding population of off-the-disk stars (ODSs). Despite their diverse kinematic properties, all three-populations appear to be massive (WR/O/B types) and have similarly limited life span $\tau_\star \sim 6-15$ Myr. Several scenarios, including star formation induced by SMBH's close encounters with one or more gas clouds as well as impulsive close scattering by a putative intermediate-mass companion (IMC) of Sgr A$^\star$ possible an intermediate-mass black hole (IMBH), have been proposed to explain piecemeal for the origin and dynamical evolution of S-stars, CWSs, ODSs, as well as hyper-velocity stars in the Galaxy. But, their coexistence and the origin of a recently discovered zone of avoidance in S-stars' eccentricity-peri-centric-distance distribution remain enigmatic. Here, we construct a unified model to comprehensively take into account these stars' interaction with each other, their single natal disk, and an independent IMC. We show their disparate present-day orbits would only be concurrently attainable, within their multi-Myr age, under the combined influence of IMC's secular perturbation and these stars' resonant relaxation in a depleting gaseous-disk environment.

astro-ph.GA↗

arXiv2026

TSAssistant: A Human-in-the-Loop Agentic Framework for Automated Target Safety Assessment

Target Safety Assessment (TSA) requires systematic integration of genetic, transcriptomic, target homology, pharmacological, and clinical data to evaluate potential safety liabilities of therapeutic targets. This process is labor-intensive and expert-dependent, posing challenges in scalability and reproducibility. We present TSAssistant, a human-in-the-loop multi-agent framework that decomposes TSA report generation into a workflow of specialized subagents: Research Subagents that each ground and cite a single TSA domain, and Synthesis Subagents that integrate findings across domains. Subagents retrieve and synthesize evidence from curated biomedical sources through standardized tool interfaces and produce individually citable, evidence-grounded sections, with behavior shaped by a hierarchical instruction architecture that separates coordination logic from domain expertise and user intent. To complement these soft constraints, programmatic execution hooks and persistent memory stores enforce hard constraints across the workflow, while an interactive refinement loop allows experts to review and revise individual sections with full conversational context preserved across iterations. Rather than a single holistic comparison, we decompose report quality into reproducibility, evidential grounding, task-level accuracy, and controllability under expert oversight, finding high reproducibility and grounding, substantial agreement with the human reference, and net-positive expert-driven refinement.

cs.CL↗

arXiv2026

Enhancement of the Rate of Tidal Disruption Events in Active Galactic Nuclei due to the Sweeping Secular Resonance Mechanism

Tidal disruption event (TDE) rates in active galactic nuclei (AGN) consistently exceed predictions from two-body relaxation, particularly in post-starburst and green valley galaxies. We explain this excess with a new mechanism: a sweeping secular resonance (SSR) driven by an intermediate-mass companion (IMC) and a depleting gaseous disk. As the disk mass declines, a resonance between stellar and IMC orbital precession sweeps through the nuclear cluster, exciting stellar eccentricities to near unity on orbital timescales far faster than gravitational relaxation. Our analytical framework, validated by N-body simulations (REBOUND), shows this mechanism requires IMC-to-SMBH mass ratios of $q \geq 10^{-3}$, disk mass ratio $p \geq 10^{-3}$, and few Myr-scale disk depletion. It is highly effective for co-orbiting IMCs but negligible for counter-orbiting ones. The TDE rate peaks at $10^{-3}-10^{-2}$ per galaxy per year for a depletion timescale $\tau_{\rm dep} \sim 10$ Myr. Even lower-mass IMCs can produce significant enhancements with compact, long-lived disks. Our model naturally explains elevated AGN TDE rates and implies that a high TDE incidence is a potential tracer of hidden parsec-scale IMCs, offering testable predictions for future AGN monitoring.

astro-ph.GA↗

arXiv2026

Automated Identification of the Tip of the Red Giant Branch in Globular Clusters with Gaia Data

This study introduces an automated approach for identifying the tip of the red giant branch (TRGB) in globular clusters, combining astronomical data with algorithmic methods. Using a dataset of 160 globular clusters and Python scripts, we matched stellar sources with Gaia data. Our script generates color-magnitude diagrams (CMDs), and uses the local outlier factor (LOF) algorithm to remove outliers. Applying a second-degree polynomial to fit red giant branch (RGB), we identify the TRGB as the star closest to the fitted curve's endpoint. By this method, we expanded TRGB samples in global clusters to 91 with newer observational data. Our results show a decreasing trend in I-band luminosity for metallicities greater than $-$1, consistent with previous studies. The results show a robust trend fitting and the $\rm M_{I}$ of TRGB is about $-$4.02 with extremely low metallicity. Our approach enhances TRGB identification efficiency while providing valuable insights for developing automatic tools in astronomical data analysis.

astro-ph.SR↗

arXiv2026

The R2Pub Telescopes for Surveying: An Overview and Performance Evaluation of the System

The R2Pub telescope, built by the Beijing Planetarium, is a 60 cm equatorial binocular telescope located at the Daocheng site of Yunnan Observatories in China, at an altitude of about 4700 m. This paper presents an overview of the R2Pub telescope system, including its design, instrumentation, and survey capabilities, and reports an initial evaluation of its system performance. R2Pub is a prime-focus binocular system, with each optical tube covering a field of view of approximately 18 square degrees. It is designed to detect a wide range of transient and variable sources in the local universe, such as variable stars, eclipsing binaries, supernovae, gamma-ray burst afterglows, tidal disruption events, active galactic nuclei, and other unknown transients. The observatory infrastructure, including the dome, equatorial mount, optical tubes, and associated subsystems, has been fully constructed and installed, and the system has entered the commissioning phase. Benefiting from the high-altitude location, good seeing conditions, and dark sky background at the Daocheng site, performance tests during commissioning show that the R2Pub system can reach a 5-sigma limiting magnitude of about 18.7 mag in the Pan-STARRS r' band with a 60 s exposure. Ongoing observations with R2Pub are expected to contribute to studies of variable and transient phenomena and to enhance public outreach in astronomy. The binocular design enables simultaneous dual-band observations, providing instantaneous color information for transient sources and improving the classification and physical characterization of their properties and evolution.

astro-ph.IM↗

arXiv2026

A Robust Launching Mechanism for Freely-Floating Planets from Host Stars with Close-in Planets

Secular perturbations from binary stars and distant massive planets can drive cold planets onto nearly parabolic orbits with pericenter passages extremely close to their host stars. Meanwhile, short-period super-Earths are frequently observed around nearby stars. Gravitational scattering between these two distinct populations can lead to substantial orbital energy exchange, liberating some intruders from the gravitational confinement of their host systems. This process offers a robust formation channel for a subset of the abundant freely floating planet population. It may also significantly perturb the original orbits of close-in planets, induce collisional trajectories between close-in planets and their host stars, and disrupt the dynamical evolution of cold planets toward close stellar encounters.

astro-ph.EP↗

arXiv2025

DiscoVerse: Multi-Agent Pharmaceutical Co-Scientist for Traceable Drug Discovery and Reverse Translation

Pharmaceutical research and development has accumulated vast and heterogeneous archives of data. Much of this knowledge stems from discontinued programs, and reusing these archives is invaluable for reverse translation. However, in practice, such reuse is often infeasible. In this work, we introduce DiscoVerse, a multi-agent co-scientist designed to support pharmaceutical research and development at Roche. Designed as a human-in-the-loop assistant, DiscoVerse enables domain-specific queries by delivering evidence-based answers: it retrieves relevant data, links across documents, summarises key findings and preserves institutional memory. We assess DiscoVerse through expert evaluation of source-linked outputs. Our evaluation spans a selected subset of 180 molecules from Roche's research and development repositories, encompassing over 0.87 billion BPE tokens and more than four decades of research. To our knowledge, this represents the first agentic framework to be systematically assessed on real pharmaceutical data for reverse translation, enabled by authorized access to confidential archives covering the full lifecycle of drug development. Our contributions include: role-specialized agent designs aligned with scientist workflows; human-in-the-loop support for reverse translation; expert evaluation; and a large-scale demonstration showing promising decision-making insights. In brief, across seven benchmark queries, DiscoVerse achieved near-perfect recall ($\geq 0.99$) with moderate precision ($0.71-0.91$). Qualitative assessments and three real-world pharmaceutical use cases further showed faithful, source-linked synthesis across preclinical and clinical evidence.

cs.CL↗

arXiv2025

MedAgentBoard: Benchmarking Multi-Agent Collaboration with Conventional Methods for Diverse Medical Tasks

The rapid advancement of Large Language Models (LLMs) has stimulated interest in multi-agent collaboration for addressing complex medical tasks. However, the practical advantages of multi-agent collaboration approaches remain insufficiently understood. Existing evaluations often lack generalizability, failing to cover diverse tasks reflective of real-world clinical practice, and frequently omit rigorous comparisons against both single-LLM-based and established conventional methods. To address this critical gap, we introduce MedAgentBoard, a comprehensive benchmark for the systematic evaluation of multi-agent collaboration, single-LLM, and conventional approaches. MedAgentBoard encompasses four diverse medical task categories: (1) medical (visual) question answering, (2) lay summary generation, (3) structured Electronic Health Record (EHR) predictive modeling, and (4) clinical workflow automation, across text, medical images, and structured EHR data. Our extensive experiments reveal a nuanced landscape: while multi-agent collaboration demonstrates benefits in specific scenarios, such as enhancing task completeness in clinical workflow automation, it does not consistently outperform advanced single LLMs (e.g., in textual medical QA) or, critically, specialized conventional methods that generally maintain better performance in tasks like medical VQA and EHR-based prediction. MedAgentBoard offers a vital resource and actionable insights, emphasizing the necessity of a task-specific, evidence-based approach to selecting and developing AI solutions in medicine. It underscores that the inherent complexity and overhead of multi-agent collaboration must be carefully weighed against tangible performance gains. All code, datasets, detailed prompts, and experimental results are open-sourced at https://medagentboard.netlify.app/.

cs.AI↗

arXiv2025

TAMER: A Test-Time Adaptive MoE-Driven Framework for EHR Representation Learning

We propose TAMER, a Test-time Adaptive MoE-driven framework for Electronic Health Record (EHR) Representation learning. TAMER introduces a framework where a Mixture-of-Experts (MoE) architecture is co-designed with Test-Time Adaptation (TTA) to jointly mitigate the intertwined challenges of patient heterogeneity and distribution shifts in EHR modeling. The MoE focuses on latent patient subgroups through domain-aware expert specialization, while TTA enables real-time adaptation to evolving health status distributions when new patient samples are introduced. Extensive experiments across four real-world EHR datasets demonstrate that TAMER consistently improves predictive performance for both mortality and readmission risk tasks when combined with diverse EHR modeling backbones. TAMER offers a promising approach for dynamic and personalized EHR-based predictions in practical clinical settings.

cs.LG↗

arXiv2024

ColaCare: Enhancing Electronic Health Record Modeling through Large Language Model-Driven Multi-Agent Collaboration

We introduce ColaCare, a framework that enhances Electronic Health Record (EHR) modeling through multi-agent collaboration driven by Large Language Models (LLMs). Our approach seamlessly integrates domain-specific expert models with LLMs to bridge the gap between structured EHR data and text-based reasoning. Inspired by the Multidisciplinary Team (MDT) approach used in clinical settings, ColaCare employs two types of agents: DoctorAgents and a MetaAgent, which collaboratively analyze patient data. Expert models process and generate predictions from numerical EHR data, while LLM agents produce reasoning references and decision-making reports within the MDT-driven collaborative consultation framework. The MetaAgent orchestrates the discussion, facilitating consultations and evidence-based debates among DoctorAgents, simulating diverse expertise in clinical decision-making. We additionally incorporate the Merck Manual of Diagnosis and Therapy (MSD) medical guideline within a retrieval-augmented generation (RAG) module for medical evidence support, addressing the challenge of knowledge currency. Extensive experiments conducted on three EHR datasets demonstrate ColaCare's superior performance in clinical mortality outcome and readmission prediction tasks, underscoring its potential to revolutionize clinical decision support systems and advance personalized precision medicine. All code, case studies and a questionnaire are available at the project website: https://colacare.netlify.app.

cs.LG↗

arXiv2024

Assessing Adversarial Robustness of Large Language Models: An Empirical Study

Large Language Models (LLMs) have revolutionized natural language processing, but their robustness against adversarial attacks remains a critical concern. We presents a novel white-box style attack approach that exposes vulnerabilities in leading open-source LLMs, including Llama, OPT, and T5. We assess the impact of model size, structure, and fine-tuning strategies on their resistance to adversarial perturbations. Our comprehensive evaluation across five diverse text classification tasks establishes a new benchmark for LLM robustness. The findings of this study have far-reaching implications for the reliable deployment of LLMs in real-world applications and contribute to the advancement of trustworthy AI systems.

cs.CL↗

arXiv2024

ClinicRealm: Re-evaluating Large Language Models with Conventional Machine Learning for Non-Generative Clinical Prediction Tasks

Large Language Models (LLMs) are increasingly deployed in medicine. However, their utility in non-generative clinical prediction, often presumed inferior to specialized models, remains under-evaluated, leading to ongoing debate within the field and potential for misuse, misunderstanding, or over-reliance due to a lack of systematic benchmarking. Our ClinicRealm study addresses this by benchmarking 15 GPT-style LLMs, 5 BERT-style models, and 11 traditional methods on unstructured clinical notes and structured Electronic Health Records (EHR), while also assessing their reasoning, reliability, and fairness. Key findings reveal a significant shift: for clinical note predictions, leading LLMs (e.g., DeepSeek-V3.1-Think, GPT-5) in zero-shot settings now decisively outperform finetuned BERT models. On structured EHRs, while specialized models excel with ample data, advanced LLMs (e.g., GPT-5, DeepSeek-V3.1-Think) show potent zero-shot capabilities, often surpassing conventional models in data-scarce settings. Notably, leading open-source LLMs can match or exceed proprietary counterparts. These results provide compelling evidence that modern LLMs are competitive tools for non-generative clinical prediction, particularly with unstructured text and offering data-efficient structured data options, thus necessitating a re-evaluation of model selection strategies. This research should serve as an important insight for medical informaticists, AI developers, and clinical researchers, potentially prompting a reassessment of current assumptions and inspiring new approaches to LLM application in predictive healthcare.

cs.CL↗

arXiv2024

EMERGE: Enhancing Multimodal Electronic Health Records Predictive Modeling with Retrieval-Augmented Generation

The integration of multimodal Electronic Health Records (EHR) data has significantly advanced clinical predictive capabilities. Existing models, which utilize clinical notes and multivariate time-series EHR data, often fall short of incorporating the necessary medical context for accurate clinical tasks, while previous approaches with knowledge graphs (KGs) primarily focus on structured knowledge extraction. In response, we propose EMERGE, a Retrieval-Augmented Generation (RAG) driven framework to enhance multimodal EHR predictive modeling. We extract entities from both time-series data and clinical notes by prompting Large Language Models (LLMs) and align them with professional PrimeKG, ensuring consistency. In addition to triplet relationships, we incorporate entities' definitions and descriptions for richer semantics. The extracted knowledge is then used to generate task-relevant summaries of patients' health statuses. Finally, we fuse the summary with other modalities using an adaptive multimodal fusion network with cross-attention. Extensive experiments on the MIMIC-III and MIMIC-IV datasets' in-hospital mortality and 30-day readmission tasks demonstrate the superior performance of the EMERGE framework over baseline models. Comprehensive ablation studies and analysis highlight the efficacy of each designed module and robustness to data sparsity. EMERGE contributes to refining the utilization of multimodal EHR data in healthcare, bridging the gap with nuanced medical contexts essential for informed clinical predictions. We have publicly released the code at https://github.com/yhzhu99/EMERGE.

cs.CL↗

arXiv2024

Clustering of Disease Trajectories with Explainable Machine Learning: A Case Study on Postoperative Delirium Phenotypes

The identification of phenotypes within complex diseases or syndromes is a fundamental component of precision medicine, which aims to adapt healthcare to individual patient characteristics. Postoperative delirium (POD) is a complex neuropsychiatric condition with significant heterogeneity in its clinical manifestations and underlying pathophysiology. We hypothesize that POD comprises several distinct phenotypes, which cannot be directly observed in clinical practice. Identifying these phenotypes could enhance our understanding of POD pathogenesis and facilitate the development of targeted prevention and treatment strategies. In this paper, we propose an approach that combines supervised machine learning for personalized POD risk prediction with unsupervised clustering techniques to uncover potential POD phenotypes. We first demonstrate our approach using synthetic data, where we simulate patient cohorts with predefined phenotypes based on distinct sets of informative features. We aim to mimic any clinical disease with our synthetic data generation method. By training a predictive model and applying SHAP, we show that clustering patients in the SHAP feature importance space successfully recovers the true underlying phenotypes, outperforming clustering in the raw feature space. We then present a case study using real-world data from a cohort of elderly surgical patients. The results showcase the utility of our approach in uncovering clinically relevant subtypes of complex disorders like POD, paving the way for more precise and personalized treatment strategies.

cs.LG↗

arXiv2024

Two-Stage Aggregation with Dynamic Local Attention for Irregular Time Series

Irregular multivariate time series data is characterized by varying time intervals between consecutive observations of measured variables/signals (i.e., features) and varying sampling rates (i.e., recordings/measurement) across these features. Modeling time series while taking into account these irregularities is still a challenging task for machine learning methods. Here, we introduce TADA, a Two-stageAggregation process with Dynamic local Attention to harmonize time-wise and feature-wise irregularities in multivariate time series. In the first stage, the irregular time series undergoes temporal embedding (TE) using all available features at each time step. This process preserves the contribution of each available feature and generates a fixed-dimensional representation per time step. The second stage introduces a dynamic local attention (DLA) mechanism with adaptive window sizes. DLA aggregates time recordings using feature-specific windows to harmonize irregular time intervals capturing feature-specific sampling rates. Then hierarchical MLP mixer layers process the output of DLA through multiscale patching to leverage information at various scales for the downstream tasks. TADA outperforms state-of-the-art methods on three real-world datasets, including the latest MIMIC IV dataset, and highlights its effectiveness in handling irregular multivariate time series and its potential for various real-world applications.

cs.LG↗

arXiv2023

A Semantic-driven Approach for Maintenance Digitalization in the Pharmaceutical Industry

The digital transformation of pharmaceutical industry is a challenging task due to the high complexity of involved elements and the strict regulatory compliance. Maintenance activities in the pharmaceutical industry play an essential role in ensuring product quality and integral functioning of equipment and premises. This paper first identifies the key challenges of digitalization in pharmaceutical industry and creates the corresponding problem space for key involved elements. A literature review is conducted to investigate the mainstream maintenance strategies, digitalization models, tools and official guidance from authorities in pharmaceutical industry. Based on the review result, a semantic-driven digitalization framework is proposed aiming to improve the digital continuity and cohesion of digital resources and technologies for maintenance activities in the pharmaceutical industry. A case study is conducted to verify the feasibility of the proposed framework based on the water sampling activities in Merck Serono facility in Switzerland. A tool-chain is presented to enable the functional modules of the framework. Some of the key functional modules within the framework are implemented and have demonstrated satisfactory performance. As one of the outcomes, a digital sampling assistant with web-based services is created to support the automated workflow of water sampling activities. The implementation result proves the potential of the proposed framework to solve the identified problems of maintenance digitalization in the pharmaceutical industry.

eess.SY↗

arXiv2023

PRISM: Mitigating EHR Data Sparsity via Learning from Missing Feature Calibrated Prototype Patient Representations

Electronic Health Records (EHRs) contain a wealth of patient data; however, the sparsity of EHRs data often presents significant challenges for predictive modeling. Conventional imputation methods inadequately distinguish between real and imputed data, leading to potential inaccuracies of patient representations. To address these issues, we introduce PRISM, a framework that indirectly imputes data by leveraging prototype representations of similar patients, thus ensuring compact representations that preserve patient information. PRISM also includes a feature confidence learner module, which evaluates the reliability of each feature considering missing statuses. Additionally, PRISM introduces a new patient similarity metric that accounts for feature confidence, avoiding over-reliance on imprecise imputed values. Our extensive experiments on the MIMIC-III, MIMIC-IV, PhysioNet Challenge 2012, eICU datasets demonstrate PRISM's superior performance in predicting in-hospital mortality and 30-day readmission tasks, showcasing its effectiveness in handling EHR data sparsity. For the sake of reproducibility and further research, we have publicly released the code at https://github.com/yhzhu99/PRISM.

cs.LG↗

arXiv2023

The Orbital Structure and Selection Effects of the Galactic Center S-Star Cluster

The orbital distribution of the S-star cluster surrounding the supermassive black hole in the center of the Milky Way is analyzed. A tight, roughly exponential dependence of the pericenter distance r$_{p}$ on orbital eccentricity e$_{\star}$ is found, $\log ($r$_p)\sim$(1-e$_{\star}$), which cannot be explained simply by a random distribution of semi-major axes and eccentricities. No stars are found in the region with high e$_{\star}$ and large log r$_{p}$ or in the region with low e$_{\star}$ and small log r$_{p}$. G-clouds follow the same correlation. The likelihood P(log r$_p$,(1-e$_{\star}$)) to determine the orbital parameters of S-stars is determined. P is very small for stars with large e$_{\star}$ and large log r$_{p}$. S-stars might exist in this region. To determine their orbital parameters, one however needs observations over a longer time period. On the other hand, if stars would exist in the region of low log r$_{p}$ and small e$_{\star}$, their orbital parameters should by now have been determined. That this region is unpopulated therefore indicates that no S-stars exist with these orbital characteristics, providing constraints for their formation. We call this region, defined by $\log$ (r$_p$/AU) $<$ 1.57+2.6(1-e$_{\star})$, the zone of avoidance. Finally, it is shown that the observed frequency of eccentricities and pericenter distances is consistent with a random sampling of log r$_{p}$ and e$_{\star}$. However, only if one takes into account that no stars exist in the zone of avoidance and that orbital parameters cannot yet be determined for stars with large r$_{p}$ and large e$_{\star}$.

astro-ph.GA↗