arXiv subjects

Junwei Liu

Publications and source records attributed to Junwei Liu.

At least 37 records · Page 2Linked to original sources

arXiv2026

Renormalization-Inspired Effective Field Neural Networks for Scalable Modeling of Classical and Quantum Many-Body Systems

We introduce Effective Field Neural Networks (EFNNs), a new architecture based on continued functions -- mathematical tools used in renormalization to handle divergent perturbative series. Our key insight is that neural networks can implement these continued functions directly, providing a principled approach to many-body interactions. Testing on three systems (a classical 3-spin infinite- range model, a continuous classical Heisenberg spin system, and a quantum double exchange model), we find that EFNN outperforms standard deep networks, ResNet, and DenseNet. Most striking is EFNN's generalization: trained on $10 \times 10$ lattices, it accurately predicts behavior on systems up to $40\times 40$ with no additional training -- and the accuracy improves with system size, with a computational time speed-up of $10^{3}$ compared to ED for $40\times 40$ lattice. This demonstrates that EFNN captures the underlying physics rather than merely fitting data, making it valuable beyond many-body problems to any field where renormalization ideas apply.

physics.comp-ph↗

arXiv2026

Evolving Interactive Diagnostic Agents in a Virtual Clinical Environment

We present a framework for training large language models (LLMs) as diagnostic agents with reinforcement learning, enabling them to manage multi-turn interactive diagnostic processes, adaptively select examinations, and commit to final diagnoses. Unlike instruction-tuned models trained on static data, our method acquires diagnostic strategies through dynamic exploration and outcome-based feedback, mapping evolving patient states to the next optimal examination and subsequent diagnosis. Our contributions include: (i) DiagGym, a diagnostics world model trained with electronic health records, serving as a virtual clinical environment to support closed-loop in-silico training and evaluation for interactive diagnosis; (ii) DiagAgent, trained via end-to-end multi-turn RL to learn dynamic diagnostic policies that optimize both interactive effectiveness and final accuracy; (iii) DiagBench, a multi-center diagnostic benchmark designed to evaluate multi-turn diagnostic interaction trajectories. The benchmark comprises 2.2K physician-validated cases sourced from 4 distinct distributions, alongside 3.3K physician-written rubrics for granular process-oriented evaluation. (iv) Extensive evaluations demonstrate DiagAgent's superior performance across both in-domain and out-of-domain (OOD) settings. DiagAgent significantly outperforms 11 SOTA LLMs and 2 prompt-engineered agents. In the end-to-end setting, it delivers a 11.20% increase in diagnostic accuracy and a 17.58% boost in examination recommendation F1 score, while consistently maintaining SOTA performance across all three external centers. Furthermore, in rubric-based evaluations, it surpasses the next-best model by 7.1% in weighted rubric score. These findings indicate that learning policies in interactive clinical environments confers long-term diagnostic management abilities unattainable through passive training.

cs.CL↗

arXiv2026

PulseMind: A Multi-Modal Medical Model for Real-World Clinical Diagnosis

Recent advances in medical multi-modal models focus on specialized image analysis like dermatology, pathology, or radiology. However, they do not fully capture the complexity of real-world clinical diagnostics, which involve heterogeneous inputs and require ongoing contextual understanding during patient-physician interactions. To bridge this gap, we introduce PulseMind, a new family of multi-modal diagnostic models that integrates a systematically curated dataset, a comprehensive evaluation benchmark, and a tailored training framework. Specifically, we first construct a diagnostic dataset, MediScope, which comprises 98,000 real-world multi-turn consultations and 601,500 medical images, spanning over 10 major clinical departments and more than 200 sub-specialties. Then, to better reflect the requirements of real-world clinical diagnosis, we develop the PulseMind Benchmark, a multi-turn diagnostic consultation benchmark with a four-dimensional evaluation protocol comprising proactiveness, accuracy, usefulness, and language quality. Finally, we design a training framework tailored for multi-modal clinical diagnostics, centered around a core component named Comparison-based Reinforcement Policy Optimization (CRPO). Compared to absolute score rewards, CRPO uses relative preference signals from multi-dimensional com-parisons to provide stable and human-aligned training guidance. Extensive experiments demonstrate that PulseMind achieves competitive performance on both the diagnostic consultation benchmark and public medical benchmarks.

cs.CV↗

arXiv2026

MLB: A Scenario-Driven Benchmark for Evaluating Large Language Models in Clinical Applications

The proliferation of Large Language Models (LLMs) presents transformative potential for healthcare, yet practical deployment is hindered by the absence of frameworks that assess real-world clinical utility. Existing benchmarks test static knowledge, failing to capture the dynamic, application-oriented capabilities required in clinical practice. To bridge this gap, we introduce a Medical LLM Benchmark MLB, a comprehensive benchmark evaluating LLMs on both foundational knowledge and scenario-based reasoning. MLB is structured around five core dimensions: Medical Knowledge (MedKQA), Safety and Ethics (MedSE), Medical Record Understanding (MedRU), Smart Services (SmartServ), and Smart Healthcare (SmartCare). The benchmark integrates 22 datasets (17 newly curated) from diverse Chinese clinical sources, covering 64 clinical specialties. Its design features a rigorous curation pipeline involving 300 licensed physicians. Besides, we provide a scalable evaluation methodology, centered on a specialized judge model trained via Supervised Fine-Tuning (SFT) on expert annotations. Our comprehensive evaluation of 10 leading models reveals a critical translational gap: while the top-ranked model, Kimi-K2-Instruct (77.3% accuracy overall), excels in structured tasks like information extraction (87.8% accuracy in MedRU), performance plummets in patient-facing scenarios (61.3% in SmartServ). Moreover, the exceptional safety score (90.6% in MedSE) of the much smaller Baichuan-M2-32B highlights that targeted training is equally critical. Our specialized judge model, trained via SFT on a 19k expert-annotated medical dataset, achieves 92.1% accuracy, an F1-score of 94.37%, and a Cohen's Kappa of 81.3% for human-AI consistency, validating a reproducible and expert-aligned evaluation protocol. MLB thus provides a rigorous framework to guide the development of clinically viable LLMs.

cs.LG↗

arXiv2026

MedDialogRubrics: A Comprehensive Benchmark and Evaluation Framework for Multi-turn Medical Consultations in Large Language Models

Medical conversational AI (AI) plays a pivotal role in the development of safer and more effective medical dialogue systems. However, existing benchmarks and evaluation frameworks for assessing the information-gathering and diagnostic reasoning abilities of medical large language models (LLMs) have not been rigorously evaluated. To address these gaps, we present MedDialogRubrics, a novel benchmark comprising 5,200 synthetically constructed patient cases and over 60,000 fine-grained evaluation rubrics generated by LLMs and subsequently refined by clinical experts, specifically designed to assess the multi-turn diagnostic capabilities of LLM. Our framework employs a multi-agent system to synthesize realistic patient records and chief complaints from underlying disease knowledge without accessing real-world electronic health records, thereby mitigating privacy and data-governance concerns. We design a robust Patient Agent that is limited to a set of atomic medical facts and augmented with a dynamic guidance mechanism that continuously detects and corrects hallucinations throughout the dialogue, ensuring internal coherence and clinical plausibility of the simulated cases. Furthermore, we propose a structured LLM-based and expert-annotated rubric-generation pipeline that retrieves Evidence-Based Medicine (EBM) guidelines and utilizes the reject sampling to derive a prioritized set of rubric items ("must-ask" items) for each case. We perform a comprehensive evaluation of state-of-the-art models and demonstrate that, across multiple assessment dimensions, current models face substantial challenges. Our results indicate that improving medical dialogue will require advances in dialogue management architectures, not just incremental tuning of the base-model.

cs.CL↗

arXiv2026

Observation of spin-valley locked nodal lines in a quasi-2D altermagnet

The interplay among quantum degrees of freedom-spin, orbital and momentum-has emerged as a fertile ground for realizing magnetic quantum states with transformative potential for electronic and spintronic technologies. Prominent examples include ferromagnetic Weyl semimetals and antiferromagnetic axion insulators. Recently, altermagnets(AMs) have been identified as a distinct spin-splitting class of collinear antiferromagnets(AFMs), characterized by crystal symmetry that connects magnetic sublattices in real space and enforces C-paired spin-momentum locking in reciprocal space. These materials combine the advantages of nonrelativistic spin-polarization akin to FMs and vanished net-magnetization as AFMs, making them highly promising for spintronic applications. Furthermore, they introduce nontrivial spin-momentum locking spin texture as an additional degree of freedom for realizing novel quantum phases. In this work, we report the discovery of a new type of spin-valley-locked nodal line phase in the layered AM Rb-intercalated V{_2}Te{_2}O. By combining high-resolution spin and angle-resolved photoemission spectroscopy with first-principles calculations, we observe the coexistence of both spinless and spinful nodal lines near the Fermi level. Remarkably, the spinful nodal lines exhibit uniform spin polarization within each valley, while displaying opposite spin polarizations across symmetry-paired valleys-a unique feature we term spin-valley-locked nodal lines, which is exclusive to AMs. Direct measurements of out-of-plane band dispersion using a side-cleaving technique reveal the two-dimensional nature of these nodal lines. Our findings not only unveil a previously unexplored topological phase in AMs where valley-locked spin as an additional quantum character but also establish RbV{_2}Te{_2}O as a promising platform for spintronics, valleytronics, and moire-engineered quantum devices.

cond-mat.mtrl-sci↗

arXiv2025

Atomic-scale spin sensing of a 2D $d$-wave altermagnet via helical tunneling

Altermagnetism simultaneously possesses nonrelativistic spin responses and zero net magnetization, thus combining advantages of ferromagnetism and antiferromagnetism. This superiority originates from its unique dual feature, i.e., opposite-magnetic sublattices in real space and alternating spin polarization in momentum space enforced by the same crystal symmetry. Therefore, the determination of an altermagnetic order and its unique spin response inherently necessitates atomic-scale spin-resolved measurements in real and momentum spaces, an experimental milestone yet to be achieved. Here, via utilizing the helical edge (hinge) modes of a higher order topological insulator as the spin sensor, we realize spin-resolved scanning tunneling microscopy which enables us to pin down the dual-space feature of a layered $d$-wave altermagnet, KV$_2$Se$_2$O. In real space, atomic-registered mapping demonstrates the checkerboard antiferromagnetic order together with density-wave lattice modulation, and in momentum space, spin-resolved spectroscopic imaging provides a direct visualization of d-wave spin splitting of the band structure. Critically, using this new topology-guaranteed spin filter we directly reveal the unidirectional, spin-polarized quasiparticle excitations originating from the crystal symmetry-paired X and Y valleys around opposite magnetic sublattices simultaneously --the unique spin response for $d$-wave altermagnetism. Our experiments establish a solid basis for the exploration and utilization of altermagnetism in layered materials and further facilitate access to atomic-scale spin sensing and manipulating of 2D quantum materials.

cond-mat.mes-hall↗

arXiv2025

GAPS: A Clinically Grounded, Automated Benchmark for Evaluating AI Clinicians

Current benchmarks for AI clinician systems, often based on multiple-choice exams or manual rubrics, fail to capture the depth, robustness, and safety required for real-world clinical practice. To address this, we introduce the GAPS framework, a multidimensional paradigm for evaluating Grounding (cognitive depth), Adequacy (answer completeness), Perturbation (robustness), and Safety. Critically, we developed a fully automated, guideline-anchored pipeline to construct a GAPS-aligned benchmark end-to-end, overcoming the scalability and subjectivity limitations of prior work. Our pipeline assembles an evidence neighborhood, creates dual graph and tree representations, and automatically generates questions across G-levels. Rubrics are synthesized by a DeepResearch agent that mimics GRADE-consistent, PICO-driven evidence review in a ReAct loop. Scoring is performed by an ensemble of large language model (LLM) judges. Validation confirmed our automated questions are high-quality and align with clinician judgment (90% agreement, Cohen's Kappa 0.77). Evaluating state-of-the-art models on the benchmark revealed key failure modes: performance degrades sharply with increased reasoning depth (G-axis), models struggle with answer completeness (A-axis), and they are highly vulnerable to adversarial perturbations (P-axis) as well as certain safety issues (S-axis). This automated, clinically-grounded approach provides a reproducible and scalable method for rigorously evaluating AI clinician systems and guiding their development toward safer, more reliable clinical practice. The benchmark dataset GAPS-NSCLC-preview and evaluation code are publicly available at https://huggingface.co/datasets/AQ-MedAI/GAPS-NSCLC-preview and https://github.com/AQ-MedAI/MedicalAiBenchEval.

cs.CL↗

arXiv2025

Online Full ZVS Optimization for Modular Multi-Active Bridge Converter in MV PET

Multi-active bridge (MAB) converters, the core of the state-of-the-art medium-voltage power electronic transformers, can flexibly connect multiple DC ports among distributed DC grids and loads, but suffer from hard switching under conventional single phase-shift control, especially under unbalanced voltage conversion ratios and light load conditions. Although some offline methods manage to improve the efficiency through complex optimization structures, there lacks online optimization methods that are simple but effective due to the strong coupling among ports of the converter. By leveraging the time-domain model of the MAB converter under the multiple phase-shift modulation scheme, this paper simplifies the optimization process and proposes an online optimization method that can achieve full zero-voltage switching (ZVS) operation regardless of the load conditions. The proposed method has simple solutions with only voltage conversion ratios involved and can be implemented within a wide operation range without additional sensors or advanced controllers. A four-port MAB converter is constructed as the prototype. The simulation and experimental results have verified the feasibility and superiority of the proposed online strategy in achieving ZVS operation, dynamic response, and efficiency improvement.

eess.SY↗

arXiv2025

Large Language Model-Based Agents for Software Engineering: A Survey

The recent advance in Large Language Models (LLMs) has shaped a new paradigm of AI agents, i.e., LLM-based agents. Compared to standalone LLMs, LLM-based agents substantially extend the versatility and expertise of LLMs by enhancing LLMs with the capabilities of perceiving and utilizing external resources and tools. To date, LLM-based agents have been applied and shown remarkable effectiveness in Software Engineering (SE). The synergy between multiple agents and human interaction brings further promise in tackling complex real-world SE problems. In this work, we present a comprehensive and systematic survey on LLM-based agents for SE. We collect 124 papers and categorize them from two perspectives, i.e., the SE and agent perspectives. In addition, we discuss open challenges and future directions in this critical domain. The repository of this survey is at https://github.com/FudanSELab/Agent4SE-Paper-List.

cs.SE↗

arXiv2025

EHR-R1: A Reasoning-Enhanced Foundational Language Model for Electronic Health Record Analysis

Electronic Health Records (EHRs) contain rich yet complex information, and their automated analysis is critical for clinical decision-making. Despite recent advances of large language models (LLMs) in clinical workflows, their ability to analyze EHRs remains limited due to narrow task coverage and lack of EHR-oriented reasoning capabilities. This paper aims to bridge the gap, specifically, we present EHR-Ins, a large-scale, comprehensive EHR reasoning instruction dataset, comprising 300k high-quality reasoning cases and 4M non-reasoning cases across 42 distinct EHR tasks. Its core innovation is a thinking-graph-driven framework that enables to generate high-quality reasoning data at scale. Based on it, we develop EHR-R1, a series of reasoning-enhanced LLMs with up to 72B parameters tailored for EHR analysis. Through a multi-stage training paradigm, including domain adaptation, reasoning enhancement, and reinforcement learning, EHR-R1 systematically acquires domain knowledge and diverse reasoning capabilities, enabling accurate and robust EHR analysis. Lastly, we introduce EHR-Bench, a new benchmark curated from MIMIC-IV, spanning 42 tasks, to comprehensively assess reasoning and prediction across EHR scenarios. In experiments, we show that the resulting EHR-R1 consistently outperforms state-of-the-art commercial and open-source LLMs (including DeepSeek-V3 and GPT-4o), surpassing GPT-4o by over 30 points on MIMIC-Bench and achieving a 10\% higher zero-shot AUROC on EHRSHOT. Collectively, EHR-Ins, EHR-R1, and EHR-Bench have significantly advanced the development for more reliable and clinically relevant EHR analysis.

cs.CL↗

arXiv2025

Multi-Agent Deep Research: Training Multi-Agent Systems with M-GRPO

Multi-agent systems perform well on general reasoning tasks. However, the lack of training in specialized areas hinders their accuracy. Current training methods train a unified large language model (LLM) for all agents in the system. This may limit the performances due to different distributions underlying for different agents. Therefore, training multi-agent systems with distinct LLMs should be the next step to solve. However, this approach introduces optimization challenges. For example, agents operate at different frequencies, rollouts involve varying sub-agent invocations, and agents are often deployed across separate servers, disrupting end-to-end gradient flow. To address these issues, we propose M-GRPO, a hierarchical extension of Group Relative Policy Optimization designed for vertical Multi-agent systems with a main agent (planner) and multiple sub-agents (multi-turn tool executors). M-GRPO computes group-relative advantages for both main and sub-agents, maintaining hierarchical credit assignment. It also introduces a trajectory-alignment scheme that generates fixed-size batches despite variable sub-agent invocations. We deploy a decoupled training pipeline in which agents run on separate servers and exchange minimal statistics via a shared store. This enables scalable training without cross-server backpropagation. In experiments on real-world benchmarks (e.g., GAIA, XBench-DeepSearch, and WebWalkerQA), M-GRPO consistently outperforms both single-agent GRPO and multi-agent GRPO with frozen sub-agents, demonstrating improved stability and sample efficiency. These results show that aligning heterogeneous trajectories and decoupling optimization across specialized agents enhances tool-augmented reasoning tasks.

cs.AI↗

arXiv2025

Broadband nonlinear Hall response and multiple wave mixing in a room temperature altermagnet

Crystalline symmetries determine the linear and nonlinear response of materials to external stimuli such as mechanical pressure and electromagnetic fields, governing phenomena such as piezoelectricity, optical activity, and multiple wave mixing with wide ranging technological applications. Altermagnets present a new class of materials with magnetic crystalline order where specific crystal symmetry operations connect antiferromagnetic sublattices, leading to non-relativistic spin-splitting of the electronic band structure. Hence, the electric material properties of altermagnets should uniquely mirror these underlying symmetry properties, potentially giving rise to novel phenomena in response to external driving fields. Here, we report the discovery of a broadband third-order nonlinear anomalous Hall effect in altermagnetic CrSb at room temperature. The comparison of our observations with symmetry analyses and model calculations shows that this nonlinear Hall response is induced by the nonlinear electric susceptibility of a Berry curvature quadrupole, which exists within the spin-split band structure of CrSb and is characterized by the underlying crystalline and magnetic symmetries. We then utilize this third-order nonlinear electric susceptibility of CrSb to realize a multiple wave mixing device with pronounced four wave mixing output, which could, in principle, be extended to THz frequencies. Our study discovers that the crystalline magnetic order of altermagnets determines their nonlinear electric material properties, which could facilitate applications in high-frequency electronics, THz generation, communication networks, and energy harvesting.

cond-mat.mes-hall↗

arXiv2025

PE-TSFM: Self-Supervised Time-Series Learning for Generalizable Power Converter Health Monitoring under Unseen Conditions

Data-driven health monitoring of power converters remains limited by poor generalization to unseen operating conditions. This work addresses this out-of-distribution (OOD) challenge by building a domain-specific time-series foundation model (PE-TSFM) that learns representations directly from large-scale unlabeled converter data. Unlike generic TSFMs trained on broad time-series datasets, the proposed PE-TSFM is pre-trained entirely on domain data, enabling it to learn the physical relationships unique to power electronics. To further tailor the model to this domain, we introduce a dual-attention mechanism that captures both temporal patterns and inter-channel dependencies. While generic TSFMs primarily model temporal dependencies, the added channel attention captures inter-sensor physical relationships essential for converter degradation analysis. A dataset containing 141 million unlabeled timestamps from an operating power converter is used for pre-training. Experiments show that PE-TSFM achieves 92% accuracy under unseen operating conditions. In contrast, generic TSFMs achieve around 60% and conventional time-series models achieve around 40% accuracy. This result confirms the strong OOD generalization of the proposed PE-TSFM. Ablation studies further verify that the introduced channel attention mechanism significantly improves model performance. In addition, we conduct detailed studies on model scalability, hyperparameter sensitivity, and interpretability to provide a comprehensive understanding of the proposed approach.

eess.SY↗

arXiv2025

Can Agents Fix Agent Issues?

LLM-based agent systems are emerging as a new software paradigm and have been widely adopted across diverse domains such as medicine, robotics, and programming. However, maintaining these systems requires substantial effort, as they are inevitably prone to bugs and continually evolve to meet changing external requirements. Therefore, automatically resolving agent issues (i.e., bug reports or feature requests) is a crucial and challenging task. While recent software engineering (SE) agents (e.g., SWE-agent) have shown promise in addressing issues in traditional software systems, it remains unclear how effectively they can resolve real-world issues in agent systems, which differ significantly from traditional software. To fill this gap, we first manually analyze 201 real-world agent issues and identify common categories of agent issues. We then spend 500 person-hours constructing AgentIssue-Bench, a reproducible benchmark comprising 50 agent issue resolution tasks (each with an executable environment and failure-triggering tests). We further evaluate state-of-the-art SE agents on AgentIssue-Bench and reveal their limited effectiveness (i.e., with only 0.67% - 4.67% resolution rates). These results underscore the unique challenges of maintaining agent systems compared to traditional software, highlighting the need for further research to develop advanced SE agents for resolving agent issues. Data and code are available at https://github.com/alfin06/AgentIssue-Bench.

cs.AI↗

arXiv2025

Multiple Topological Phases Controlled via Strain in Two-Dimensional Altermagnets

Altermagnets (AMs) are an emergent class of magnetic materials that combine properties of ferromagnets and antiferromagnets, exhibiting spin-polarized Fermi surfaces and zero net magnetic moment due to combined time-reversal and crystal symmetry. Here, we construct a Kondo-lattice model on a two-dimensional square Lieb lattice to investigate the topological properties of AMs. We identify a type-II quantum spin Hall state characterized by spin-polarized counterpropagating edge states. Breaking the $C_{4z}\mathcal{T}$ symmetry, which connects magnetic sublattices, induces a transition to a quantum anomalous Hall state. We further establish a strain-induced mechanism to control these topological phase transitions and present the corresponding phase diagram. Finally, we demonstrate the predicted transitions in monolayer CrO, a realistic altermagnetic candidate, using first-principles calculations. Our findings highlight the potential of 2D AMs as a versatile platform for topological spintronics, enabling strain-tunable helical and chiral edge states within a single system.

cond-mat.mes-hall↗

arXiv2025

MedResearcher-R1: Expert-Level Medical Deep Researcher via A Knowledge-Informed Trajectory Synthesis Framework

Recent developments in Large Language Model (LLM)-based agents have shown impressive capabilities spanning multiple domains, exemplified by deep research systems that demonstrate superior performance on complex information-seeking and synthesis tasks. While general-purpose deep research agents have shown impressive capabilities, they struggle significantly with medical domain challenges, as evidenced by leading proprietary systems achieving limited accuracy on complex medical benchmarks. The key limitations are: (1) the model lacks sufficient dense medical knowledge for clinical reasoning, and (2) the framework is constrained by the absence of specialized retrieval tools tailored for medical contexts. We present a medical deep research agent that addresses these challenges through two core innovations. First, we develop a novel data synthesis framework using medical knowledge graphs, extracting the longest chains from subgraphs around rare medical entities to generate complex multi-hop question-answer pairs. Second, we integrate a custom-built private medical retrieval engine alongside general-purpose tools, enabling accurate medical information synthesis. Our approach generates 2100+ diverse trajectories across 12 medical specialties, each averaging 4.2 tool interactions. Through a two-stage training paradigm combining supervised fine-tuning and online reinforcement learning with composite rewards, our MedResearcher-R1-32B model demonstrates exceptional performance, establishing new state-of-the-art results on medical benchmarks while maintaining competitive performance on general deep research tasks. Our work demonstrates that strategic domain-specific innovations in architecture, tool design, and training data construction can enable smaller open-source models to outperform much larger proprietary systems in specialized domains.

cs.CL↗

arXiv2025

Quantum-Boosted High-Fidelity Deep Learning

A fundamental limitation of probabilistic deep learning is its predominant reliance on Gaussian priors. This simplistic assumption prevents models from accurately capturing the complex, non-Gaussian landscapes of natural data, particularly in demanding domains like complex biological data, severely hindering the fidelity of the model for scientific discovery. The physically-grounded Boltzmann distribution offers a more expressive alternative, but it is computationally intractable on classical computers. To date, quantum approaches have been hampered by the insufficient qubit scale and operational stability required for the iterative demands of deep learning. Here, we bridge this gap by introducing the Quantum Boltzmann Machine-Variational Autoencoder (QBM-VAE), a large-scale and long-time stable hybrid quantum-classical architecture. Our framework leverages a quantum processor for efficient sampling from the Boltzmann distribution, enabling its use as a powerful prior within a deep generative model. Applied to million-scale single-cell datasets from multiple sources, the QBM-VAE generates a latent space that better preserves complex biological structures, consistently outperforming conventional Gaussian-based deep learning models like VAE and SCVI in essential tasks such as omics data integration, cell-type classification, and trajectory inference. It also provides a typical example of introducing a physics priori into deep learning to drive the model to acquire scientific discovery capabilities that breaks through data limitations. This work provides the demonstration of a practical quantum advantage in deep learning on a large-scale scientific problem and offers a transferable blueprint for developing hybrid quantum AI models.

cs.LG↗