Searcharxiv⌕ Search

arXiv subjects

Chen Zhao

Publications and source records attributed to Chen Zhao.

At least 109 records · Page 6Linked to original sources
arXiv2026

Does FLUX Already Know How to Perform Physically Plausible Image Composition?

Image composition aims to seamlessly insert a user-specified object into a new scene, but existing models struggle with complex lighting (e.g., accurate shadows, water reflections) and diverse, high-resolution inputs. Modern text-to-image diffusion models (e.g., SD3.5, FLUX) already encode essential physical and resolution priors, yet lack a framework to unleash them without resorting to latent inversion, which often locks object poses into contextually inappropriate orientations, or brittle attention surgery. We propose SHINE, a training-free framework for Seamless, High-fidelity Insertion with Neutralized Errors. SHINE introduces manifold-steered anchor loss, leveraging pretrained customization adapters (e.g., IP-Adapter) to guide latents for faithful subject representation while preserving background integrity. Degradation-suppression guidance and adaptive background blending are proposed to further eliminate low-quality outputs and visible seams. To address the lack of rigorous benchmarks, we introduce ComplexCompo, featuring diverse resolutions and challenging conditions such as low lighting, strong illumination, intricate shadows, and reflective surfaces. Experiments on ComplexCompo and DreamEditBench show state-of-the-art performance on standard metrics (e.g., DINOv2) and human-aligned scores (e.g., DreamSim, ImageReward, VisionReward). Code is available at https://github.com/ZhumingLian/SHINE.

cs.CV↗
arXiv2026

Improving Generalizability of Hip Fracture Risk Prediction via Domain Adaptation Across Multiple Cohorts

Clinical risk prediction models often fail to be generalized across cohorts because underlying data distributions differ by clinical site, region, demographics, and measurement protocols. This limitation is particularly pronounced in hip fracture risk prediction, where the performance of models trained on one cohort (the source cohort) can degrade substantially when deployed in other cohorts (target cohorts). We used a shared set of clinical and DXA-derived features across three large cohorts - the Study of Osteoporotic Fractures (SOF), the Osteoporotic Fractures in Men Study (MrOS), and the UK Biobank (UKB), to systematically evaluate the performance of three domain adaptation methods - Maximum Mean Discrepancy (MMD), Correlation Alignment (CORAL), and Domain - Adversarial Neural Networks (DANN) and their combinations. For a source cohort with males only and a source cohort with females only, domain-adaptation methods consistently showed improved performance than the no-adaptation baseline (source-only training), and the use of combinations of multiple domain adaptation methods delivered the largest and most stable gains. The method that combines MMD, CORAL, and DANN achieved the highest discrimination with the area under curve (AUC) of 0.88 for a source cohort with males only and 0.95 for a source cohort with females only), demonstrating that integrating multiple domain adaptation methods could produce feature representations that are less sensitive to dataset differences. Unlike existing methods that rely heavily on supervised tuning or assume known outcomes of samples in target cohorts, our outcome-free approaches enable the model selection under realistic deployment conditions and improve generalization of models in hip fracture risk prediction.

cs.LG↗
arXiv2026

RPDR: A Round-trip Prediction-Based Data Augmentation Framework for Long-Tail Question Answering

Long-tail question answering presents significant challenges for large language models (LLMs) due to their limited ability to acquire and accurately recall less common knowledge. Retrieval-augmented generation (RAG) systems have shown great promise in mitigating this limitation by integrating external retrieval mechanisms. However, dense retrieval models often face the same difficulties when generalizing to rare or niche knowledge. In this study, we introduce RPDR, a novel data augmentation framework that selects high-quality easy-to-learn training data, to enhance dense retrievers. Our approach is built around three core components: synthetic data generation, data selection with Round-Trip prediction to identify easy-to-learn instances, and retriever training with these instances. We evaluate RPDR on two long-tail retrieval benchmarks, PopQA and EntityQuestion, demonstrating substantial improvements over existing retrievers like BM25 and Contriver, especially on extremely long-tail categories. We identify the strengths and limitations of RPDR through detailed human analysis and propose a dynamic routing mechanism to dynamically route queries to specialized retrieval modules to further improve retrieval performance.

cs.CL↗
arXiv2026

MRMR: A Realistic and Expert-Level Multidisciplinary Benchmark for Reasoning-Intensive Multimodal Retrieval

We introduce MRMR, the first expert-level multidisciplinary multimodal retrieval benchmark requiring intensive reasoning. MRMR contains 1,502 queries spanning 23 domains, with positive documents carefully verified by human experts. Compared to prior benchmarks, MRMR introduces three key advancements. First, it challenges retrieval systems across diverse areas of expertise, enabling fine-grained model comparison across domains. Second, queries are reasoning-intensive, with images requiring deeper interpretation such as diagnosing microscopic slides. We further introduce Contradiction Retrieval, a novel task requiring models to identify conflicting concepts. Finally, queries and documents are constructed as image-text interleaved sequences. Unlike earlier benchmarks restricted to single images or unimodal documents, MRMR offers a realistic setting with multi-image queries and mixed-modality corpus documents. We conduct an extensive evaluation of 4 categories of multimodal retrieval systems and 14 frontier models on MRMR. The text embedding model Qwen3-Embedding with LLM-generated image captions achieves the highest performance, highlighting substantial room for improving multimodal retrieval models. Although latest multimodal models such as Ops-MM-Embedding perform competitively on expert-domain queries, they fall short on reasoning-intensive tasks. We believe that MRMR paves the way for advancing multimodal retrieval in more realistic and challenging scenarios.

cs.IR↗
arXiv2026

Learning A Physical-aware Diffusion Model Based on Transformer for Underwater Image Enhancement

Underwater visuals undergo various complex degradations, inevitably influencing the efficiency of underwater vision tasks. Recently, diffusion models were employed to underwater image enhancement (UIE) tasks, and gained SOTA performance. However, these methods fail to consider the physical properties and underwater imaging mechanisms in the diffusion process, limiting information completion capacity of diffusion models. In this paper, we introduce a novel UIE framework, named PA-Diff, designed to exploiting the knowledge of physics to guide the diffusion process. PA-Diff consists of Physics Prior Generation (PPG) Branch, Implicit Neural Reconstruction (INR) Branch, and Physics-aware Diffusion Transformer (PDT) Branch. Our designed PPG branch aims to produce the prior knowledge of physics. With utilizing the physics prior knowledge to guide the diffusion process, PDT branch can obtain underwater-aware ability and model the complex distribution in real-world underwater scenes. INR Branch can learn robust feature representations from diverse underwater image via implicit neural representation, which reduces the difficulty of restoration for PDT branch. Extensive experiments prove that our method achieves best performance on UIE tasks.

cs.CV↗
arXiv2026

Data Science and Technology Towards AGI Part I: Tiered Data Management

The development of artificial intelligence can be viewed as an evolution of data-driven learning paradigms, with successive shifts in data organization and utilization continuously driving advances in model capability. Current LLM research is dominated by a paradigm that relies heavily on unidirectional scaling of data size, increasingly encountering bottlenecks in data availability, acquisition cost, and training efficiency. In this work, we argue that the development of AGI is entering a new phase of data-model co-evolution, in which models actively guide data management while high-quality data, in turn, amplifies model capabilities. To implement this vision, we propose a tiered data management framework, designed to support the full LLM training lifecycle across heterogeneous learning objectives and cost constraints. Specifically, we introduce an L0-L4 tiered data management framework, ranging from raw uncurated resources to organized and verifiable knowledge. Importantly, LLMs are fully used in data management processes, such as quality scoring and content editing, to refine data across tiers. Each tier is characterized by distinct data properties, management strategies, and training roles, enabling data to be strategically allocated across LLM training stages, including pre-training, mid-training, and alignment. The framework balances data quality, acquisition cost, and marginal training benefit, providing a systematic approach to scalable and sustainable data management. We validate the effectiveness of the proposed framework through empirical studies, in which tiered datasets are constructed from raw corpora and used across multiple training phases. Experimental results demonstrate that tier-aware data utilization significantly improves training efficiency and model performance. To facilitate further research, we release our tiered datasets and processing tools to the community.

cs.AI↗
arXiv2026

Theory of Integer Quantum Hall Effect in Irrational Magnetic Field

The conventional theory of the integer quantum Hall effect (IQHE) fails for irrational magnetic fields owing to the breakdown of magnetic translational symmetry. Here, based on the recently proposed incommensurate energy band (IEB) theory, we present a universal IQHE theory that does not rely on magnetic translation symmetry and is applicable to both rational and irrational magnetic fluxes. Using the square lattice as a paradigmatic example, we first show that the IEB framework provides a superior description of its energy spectrum in a magnetic field, as it explicitly reveals the momentum-space distribution of eigenstates. Key to our IQHE theory is that each gap in the IEB spectrum is intrinsically labeled by an integer pair (m,g), defined by the corresponding Bragg planes. When the Fermi energy lies within such a gap, the occupied electron states $N_{\text{occ}}$ is determined by the k-space volume enclosed by these Bragg planes, leading to the fundamental relation $N_{\text{occ}}/N_0 = m(ϕ/ϕ_0) + g$. Through Středa formula, this leads directly to the quantized Hall conductance $σ_{xy} = m e^2/h$ under arbitrary magnetic fields. Our work resolves the long-standing problem of IQHE under irrational flux, and establishes a new paradigm for IQHE.

cond-mat.mes-hall↗
arXiv2026

SAGE: Benchmarking and Improving Retrieval for Deep Research Agents

Deep research agents have emerged as powerful systems for addressing complex queries. Meanwhile, LLM-based retrievers have demonstrated strong capability in following instructions or reasoning. This raises a critical question: can LLM-based retrievers effectively contribute to deep research agent workflows? To investigate this, we introduce SAGE, a benchmark for scientific literature retrieval comprising 1,200 queries across four scientific domains, with a 200,000 paper retrieval corpus. We evaluate six deep research agents and find that all systems struggle with reasoning-intensive retrieval. Using DR Tulu as backbone, we further compare BM25 and LLM-based retrievers (i.e., ReasonIR and gte-Qwen2-7B-instruct) as alternative search tools. Surprisingly, BM25 significantly outperforms LLM-based retrievers by approximately 30%, as existing agents generate keyword-oriented sub-queries. To improve performance, we propose a corpus-level test-time scaling framework that uses LLMs to augment documents with metadata and keywords, making retrieval easier for off-the-shelf retrievers. This yields 8% and 2% gains on short-form and open-ended questions, respectively.

cs.IR↗
arXiv2026

Factuality on Demand: Controlling the Factuality-Informativeness Trade-off in Text Generation

Large language models (LLMs) encode knowledge with varying degrees of confidence. When responding to queries, models face an inherent trade-off: they can generate responses that are less informative but highly factual, or more informative but potentially less accurate. Different applications demand different balances between informativeness and factuality. We introduce Factuality-Controlled Generation (FCG), a framework that enables users to specify factuality constraints alongside their queries. We propose to evaluate FCG performance on two dimensions: adherence to factuality constraints and response informativeness. We propose to train models on the FCG task using synthetic data, and show that our synthetic training significantly improves models' ability to both respect factuality requirements and maintain informativeness in their outputs.

cs.CL↗
arXiv2026

SciArena: An Open Evaluation Platform for Non-Verifiable Scientific Literature-Grounded Tasks

We present SciArena, an open and collaborative platform for evaluating foundation models on scientific literature-grounded tasks. Unlike traditional benchmarks for scientific literature understanding and synthesis, SciArena engages the research community directly, following the Chatbot Arena evaluation approach of community voting on model comparisons. By leveraging collective intelligence, SciArena offers a community-driven evaluation of model performance on open-ended scientific tasks that demand literature-grounded, long-form responses. The platform currently supports 47 foundation models and has collected over 20,000 votes from human researchers across diverse scientific domains. Our analysis of the data collected so far confirms its high quality. We discuss the results and insights based on the model ranking leaderboard. To further promote research in building model-based automated evaluation systems for literature tasks, we release SciArena-Eval, a meta-evaluation benchmark based on collected preference data. It measures the accuracy of models in judging answer quality by comparing their pairwise assessments with human votes. Our experiments highlight the benchmark's challenges and emphasize the need for more reliable automated evaluation methods.

cs.CL↗
arXiv2026

Theory of Correlated Hofstadter Spectrum in Magic-Angle Graphene

The magnetic-field-induced correlated Chern insulator (CCI) states in magic-angle twisted bilayer graphene (MATBG) have been intensively studied in experiments, but a simple and clear understanding of their origin is still lacking. Here, we propose a unified theoretical framework for the CCI states in MATBG that successfully explains most experimental observations. The key insight of our theory is that, due to the very narrow bandwidth of MATBG, correlation-enhanced valley and spin Zeeman terms are critical for shaping the intricate Hofstadter spectrum, resulting in an interwoven, flavor-resolved (spin and valley) Hofstadter spectrum that can well describe the observed CCI states. Crucially, due to the Zeeman effect, the crossings between these flavor-polarized Hofstadter spectra are magnetic-field-dependent, causing certain CCI states to emerge only above a critical field. This is the main mechanism underlying the critical field phenomenon of the CCI states observed in experiments. Our theory provides a clear and unified physical picture for the correlated Hofstadter spectrum in MATBG.

cond-mat.mes-hall↗
arXiv2026

Beyond Single-shot Writing: Deep Research Agents are Unreliable at Multi-turn Report Revision

Existing benchmarks for Deep Research Agents (DRAs) treat report generation as a single-shot writing task, which fundamentally diverges from how human researchers iteratively draft and revise reports via self-reflection or peer feedback. Whether DRAs can reliably revise reports with user feedback remains unexplored. We introduce Mr Dre, an evaluation suite that establishes multi-turn report revision as a new evaluation axis for DRAs. Mr Dre consists of (1) a unified long-form report evaluation protocol spanning comprehensiveness, factuality, and presentation, and (2) a human-verified feedback simulation pipeline for multi-turn revision. Our analysis of five diverse DRAs reveals a critical limitation: while agents can address most user feedback, they also regress on 16-27% of previously covered content and citation quality. Over multiple revision turns, even the best-performing agents leave significant headroom, as they continue to disrupt content outside the feedback's scope and fail to preserve earlier edits. We further show that these issues are not easily resolvable through inference-time fixes such as prompt engineering and a dedicated sub-agent for report revision.

cs.CL↗
arXiv2026

Analyzing Diffusion and Autoregressive Vision Language Models in Multimodal Embedding Space

Embedding models are a fundamental component of modern AI systems such as semantic search and retrieval-augmented generation. Recent advances in large foundation models have substantially accelerated the development of embedding models, including those based on Large Language Models (LLMs), Vision Language Models (VLMs), and Multimodal LLMs. More recently, Large Diffusion Language Models (dLLMs) and Multimodal dLLMs have emerged as competitive alternatives to autoregressive models, offering advantages such as bidirectional attention and parallel generation. This progress naturally raises a critical yet unexplored question: can Multimodal dLLMs serve as effective multimodal embedding models? To answer this, we present the first systematic study of converting Multimodal dLLMs into embedding models. We evaluate state-of-the-art Multimodal dLLMs and Autoregressive VLMs across three categories of embedding tasks: classification, visual question answering, and information retrieval. Our results show that Multimodal dLLM embeddings generally underperform their autoregressive VLM counterparts. The stronger diffusion-based model, LaViDa, lags by only 3.5 points on classification, 2.5 points on VQA, and 4.4 points on retrieval tasks, whereas the other diffusion-based model, MMaDA, exhibits substantially larger performance gaps, exceeding 20 points across all tasks. Further analysis reveals insufficient image-text alignment in diffusion-based models, accounting for the observed limitations in their embedding performance.

cs.MM↗
arXiv2026

New energy conversion system based on charge-exchange and inner-shell electron transitions

The rapidly growing demand for compact, high-energy power sources has outpaced the capabilities of conventional electrochemical systems that rely on outer-shell redox reactions. In this work, we present a new energy platform that utilizes inner-shell electron transitions that are previously inaccessible due to their high energy thresholds. By leveraging charge exchange processes between bare argon ions (Ar^18+) and neutral helium atoms, we provide clear evidence for the emission of soft X-ray and extreme-ultraviolet photons across a broad spectra range, resulting from inner-shell electron capture and cascade de-excitation. This strategy overcomes the limitations of radiative recombination by enhancing photon energy utilization through broader emission profiles more compatible with practical energy converters. Our design of a helium-filled chamber design enables precise control of output via pressure tuning, achieving a remarkable radiation power density of 6.29*10^8 W L^-1 and an unprecedented energy density of 2.64*10^6 Wh kg^-1. These results may provide a new and effective paradigm for energy conversion systems with ultra-high power and energy densities based on inner-shell electrons.

physics.atom-ph↗
arXiv2026

Histopathology-centered Computational Evolution of Spatial Omics: Integration, Mapping, and Foundation Models

Spatial omics (SO) technologies enable spatially resolved molecular profiling, while hematoxylin and eosin (H&E) imaging remains the gold standard for morphological assessment in clinical pathology. Recent computational advances increasingly place H&E images at the center of SO analysis, bridging morphology with transcriptomic, proteomic, and other spatial molecular modalities, and pushing resolution toward the single-cell level. In this survey, we systematically review the computational evolution of SO from a histopathology-centered perspective and organize existing methods into three paradigms: integration, which jointly models paired multimodal data; mapping, which infers molecular profiles from H&E images; and foundation models, which learn generalizable representations from large-scale spatial datasets. We analyze how the role of H&E images evolves across these paradigms from spatial context to predictive anchor and ultimately to representation backbone in response to practical constraints such as limited paired data and increasing resolution demands. We further summarize actionable modeling directions enabled by current architectures and delineate persistent gaps driven by data, biology, and technology that are unlikely to be resolved by model design alone. Together, this survey provides a histopathology-centered roadmap for developing and applying computational frameworks in SO.

q-bio.GN↗
arXiv2026

A Large Scale Empirical Analysis on the Adherence Gap between Standards and Tools in SBOM

A Software Bill of Materials (SBOM) is a machine-readable artifact that systematically organizes software information, enhancing supply chain transparency and security. To facilitate the exchange and utilization of SBOMs, organizations such as the Linux Foundation and OWASP have proposed SBOM standards. Following standards, organizations have developed tools for generating and utilizing SBOMs. However, limited research has examined the adherence of these SBOM tools to standard specifications, a gap that could lead to compliance failures and disruptions in SBOM utilization. This paper presents the first large-scale, two-stage empirical analysis of the adherence gap, using our automated evaluation framework, SAP. The evaluation, comprising a baseline evaluation and a one-year longitudinal follow-up, covers 55,444 SBOMs generated by six SBOM tools from 3,287 real-world repositories. Our analysis reveals persistent, fundamental limitations in current SBOM tools: (1) inadequate compliance support with policy requirements; (2) poor tool consistencies, including inter-tool consistency rates as low as 7.84% to 12.77% for package detection across languages, and significant longitudinal inconsistency, where tools show low consistency with their own prior versions; and (3) mediocre to poor accuracy for detailed software information, e.g., accuracy of package licenses below 20%. We analyze the root causes of these gaps and provide practical solutions. All the code, replication docker image, evaluation results are open sourced at [GitHub](https://github.com/dw763j/SAP) and [Zenodo](https://doi.org/10.5281/zenodo.14998624) for further researches.

cs.SE↗
arXiv2026

YOLO-IOD: Towards Real Time Incremental Object Detection

Current methods for incremental object detection (IOD) primarily rely on Faster R-CNN or DETR series detectors; however, these approaches do not accommodate the real-time YOLO detection frameworks. In this paper, we first identify three primary types of knowledge conflicts that contribute to catastrophic forgetting in YOLO-based incremental detectors: foreground-background confusion, parameter interference, and misaligned knowledge distillation. Subsequently, we introduce YOLO-IOD, a real-time Incremental Object Detection (IOD) framework that is constructed upon the pretrained YOLO-World model, facilitating incremental learning via a stage-wise parameter-efficient fine-tuning process. Specifically, YOLO-IOD encompasses three principal components: 1) Conflict-Aware Pseudo-Label Refinement (CPR), which mitigates the foreground-background confusion by leveraging the confidence levels of pseudo labels and identifying potential objects relevant to future tasks. 2) Importancebased Kernel Selection (IKS), which identifies and updates the pivotal convolution kernels pertinent to the current task during the current learning stage. 3) Cross-Stage Asymmetric Knowledge Distillation (CAKD), which addresses the misaligned knowledge distillation conflict by transmitting the features of the student target detector through the detection heads of both the previous and current teacher detectors, thereby facilitating asymmetric distillation between existing and newly introduced categories. We further introduce LoCo COCO, a more realistic benchmark that eliminates data leakage across stages. Experiments on both conventional and LoCo COCO benchmarks show that YOLO-IOD achieves superior performance with minimal forgetting.

cs.CV↗
arXiv2025

Anatomy Guided Coronary Artery Segmentation from CCTA Using Spatial Frequency Joint Modeling

Accurate coronary artery segmentation from coronary computed tomography angiography is essential for quantitative coronary analysis and clinical decision support. Nevertheless, reliable segmentation remains challenging because of small vessel calibers, complex branching, blurred boundaries, and myocardial interference. We propose a coronary artery segmentation framework that integrates myocardial anatomical priors, structure aware feature encoding, and three dimensional wavelet inverse wavelet transformations. Myocardial priors and residual attention based feature enhancement are incorporated during encoding to strengthen coronary structure representation. Wavelet inverse wavelet based downsampling and upsampling enable joint spatial frequency modeling and preserve multi scale structural consistency, while a multi scale feature fusion module integrates semantic and geometric information in the decoding stage. The model is trained and evaluated on the public ImageCAS dataset using a 3D overlapping patch based strategy with a 7:1:2 split for training, validation, and testing. Experimental results demonstrate that the proposed method achieves a Dice coefficient of 0.8082, Sensitivity of 0.7946, Precision of 0.8471, and an HD95 of 9.77 mm, outperforming several mainstream segmentation models. Ablation studies further confirm the complementary contributions of individual components. The proposed method enables more stable and consistent coronary artery segmentation under complex geometric conditions, providing reliable segmentation results for subsequent coronary structure analysis tasks.

cs.CV↗