arXiv subjects

Zeyu Gao

Publications and source records attributed to Zeyu Gao.

At least 19 recordsLinked to original sources

arXiv2026

Massive Galaxy Halos Contain Less Inner Dark Matter Than Predicted

The mass profiles of galaxy halos encode how baryons reshape dark matter distribution, yet direct observational constraints across the full radial range remain scarce. Here we combine stellar kinematics from MaNGA, H I dynamical measurements from ALFALFA, and independently calibrated halo masses of SDSS groups to statistically reconstruct the mass distribution of central galaxies over nearly two orders of magnitude in radius. We demonstrate that H I data alone do not provide reliable total halo mass estimates, necessitating an independent group-based halo-mass scale. Compared to the IllustrisTNG and EAGLE simulations, the observational profiles of low-mass halos are broadly consistent; in contrast, massive observed halos exhibit systematically lower dynamical masses at the H I radius, lower inner dark-matter masses, and lower central dark-matter fractions (about 4$σ$ difference in units of population scatter) at fixed total halo mass. After subtracting baryonic contributions, the inferred dark-matter profiles remain broadly consistent with an NFW form, but with lower effective concentrations than predicted for massive halos. These results suggest that the inner dark-matter content of massive halos has been reduced more significantly than predicted by current hydrodynamical simulations, plausibly due to long-term baryonic halo heating in massive systems.

astro-ph.GA↗

arXiv2026

Reconstructing the Projected Dark Matter Field across 0.1-100 Mpc Scales from the SDSS Survey

Dark matter sets the gravitational environment in which galaxies form and evolve, but cannot be observed directly. We present a conditional diffusion model that reconstructs the projected dark matter density field from the galaxy stellar-mass density field for direct application to galaxy surveys. The model is trained on CAMELS and validated on the independent IllustrisTNG300-1 simulation. Halo masses inferred from the reconstructed projected-aperture measurements agree well with the corresponding true values, with a scatter below 0.2 dex. On 100 kpc scales, reconstructed surface densities show a typical scatter of ~0.3 dex in the regime most relevant for observations. We apply the model to SDSS galaxies with $M_\star\ge10^9\,M_\odot$ in a contiguous low-redshift region. Averaging over 100 stochastic realizations, we reconstruct and publicly release a projected dark matter field covering $90\times90\,(h^{-1}\mathrm{Mpc})^2$ with a pixel size of $0.097\,h^{-1}\mathrm{Mpc}$. This pixel area corresponds to the characteristic projected area of halos with masses of ~$10^{10.6}\,h^{-1}\,M_\odot$. The map reveals the multiscale projected cosmic web, including cluster-scale overdensities, filaments and voids. Projected-aperture masses are statistically consistent with SDSS group-catalog masses, while the derived halo mass function broadly matches mock-catalog expectations. The reconstructed projected potential places Coma in one of the deepest wells and near a convergence region of the inferred projected acceleration field, suggesting that the reconstruction retains both local overdensities and coherent large-scale projected gravitational structure. This work shows that diffusion-based dark matter reconstruction can be applied to real galaxy surveys, enabling halo-mass- and spatially resolved dark-matter-environment-based studies of galaxy evolution in SDSS and future wide-area surveys.

astro-ph.GA↗

arXiv2026

No Detectable One-halo Galactic Conformity Signal with Halo-mass Estimates Consistent with Weak-lensing Constraints

One-halo galactic conformity is the tendency for satellites in halos with quenched centrals to have lower star-formation activity than those in halos with star-forming centrals at fixed halo mass. It is an important probe of the galaxy--halo connection and halo-wide quenching processes that may couple central and satellite evolution. However, its existence remains controversial, because conformity must be measured at fixed halo mass, while halo masses are difficult to estimate accurately. In this Letter, we measure one-halo conformity in SDSS using five stellar-mass-complete samples and three halo-mass estimates: an ML estimate whose star-forming and quenched stellar mass--halo mass relations (SHMRs) agree with independent weak-lensing constraints, and two conventional abundance-matching (AM) estimates. We quantify conformity as the difference in median $\log({\rm sSFR})$ between satellites of star-forming and quenched centrals, using both satellite-level and halo-level statistics. The two AM estimates produce strong positive conformity signals, consistent with previous AM-based measurements, but these signals are not reproduced with the ML halo masses. For the halo-level statistic, the representative AM-based signals are $+0.38\pm0.04$ dex and $+0.23\pm0.04$ dex for the luminosity-ranking and mass-ranking AM halo masses, detected relative to no conformity at about $10σ$ and $6σ$, respectively. In contrast, the ML result is consistent with no conformity, $+0.00\pm0.03$ dex; the satellite-level statistic gives a similar result. Thus, with halo-mass estimates consistent with weak-lensing constraints, we find no detectable one-halo conformity signal in the present SDSS sample, suggesting that the strong AM-based signal is largely driven by halo-mass estimation biases.

astro-ph.GA↗

arXiv2026

Data-driven registration and modeling of brain deformation for image-guided neurosurgery

Accurate compensation of brain deformation is critical for reliable image-guided neurosurgery. Surgical manipulation and tumor resection induce tissue motion, causing preoperative planning images to become misaligned with the intraoperative anatomy. In this review, we examine data-driven methods developed between 2020 and 2025 for brain deformation registration and modeling, with a particular focus on learning-based approaches. A comprehensive literature search was conducted in PubMed, IEEE Xplore, Scopus, and Web of Science using predefined inclusion and exclusion criteria for computational methods addressing brain deformation in neurosurgical imaging, resulting in 46 eligible studies. We provide a unified analysis of methodological strategies, including deep learning-based image registration, direct deformation field regression, synthesis-driven multimodal alignment, resection-aware architectures for handling missing correspondences, and hybrid models integrating biomechanical priors. We also examine dataset utilization, evaluation metrics, validation protocols, and the assessment of uncertainty and generalization across studies. While learning-based methods demonstrate promising accuracy and computational efficiency, current approaches remain limited by out-of-distribution robustness, standardized benchmarking, interpretability, and readiness for clinical deployment. Our review highlights these gaps and outlines future directions toward more robust, generalizable, and clinically translatable solutions for neurosurgical guidance. By organizing recent advances and critically assessing evaluation practices, this work provides a comprehensive reference for researchers and clinicians working on data-driven registration and modeling of brain deformation.

eess.IV↗

arXiv2026

Thinking in Scales: Accelerating Gigapixel Pathology Image Analysis via Adaptive Continuous Reasoning

Traditional whole slide image (WSI) analysis methods typically rely on the multiple instance learning (MIL) paradigm, which extracts patch-level features at high magnification and aggregates them for slide-level prediction. However, such exhaustive patch-level processing is computationally expensive, severely limiting the efficiency and scalability of WSI analysis. To address this challenge, we propose PathCTM (a Pathology-oriented Continuous Thought Model) that enables token-efficient scale-space continuous reasoning for gigapixel WSIs. PathCTM formulates diagnostic inference as a dynamic sequential information pursuit. It progressively transitions from low-magnification global to high-magnification local inspection, and adaptively terminates inference when sufficient evidence is gathered to effectively bound decision uncertainty. Specifically, it uses conditional computation for dynamic scale switching with attention-guided region pruning, coupled with confidence-aware early stopping. Extensive experiments demonstrate that, compared with standard MIL-based methods, PathCTM reduces the number of required image patches by 95.95% and shortens inference time by approximately 95.62%, while maintaining AUC without degradation. Code is available at https://github.com/JSGe-AI/PathCTM.

cs.CV↗

arXiv2026

From Patches to Evidence Balls: Class-Conditioned Evidence Retrieval for Few-Shot Whole Slide Image Classification

Whole slide image (WSI) classification is an evidence-driven task, where diagnostic cues are often sparse, spatially organized, and class-dependent. Existing MIL and vision-language methods aggregate a large pool of patch features into a single global slide representation. Under few-shot supervision, limited slide-level labels make it difficult to learn a reliable aggregation mechanism that organizes sparse local cues into compact and coherent diagnostic evidence. Moreover, a shared slide representation compresses evidence supporting a candidate class and its alternatives into the same feature, limiting class-specific reasoning and interpretability. To address these issues, we propose EviBall, a class-conditioned evidence retrieval framework for few-shot WSI classification. EviBall organizes local patches into Evidence Balls through semantic-spatial assignment and center refinement, yielding compact and spatially coherent evidence units under weak supervision. It then uses task-specific class queries, including language-guided queries for morphology-oriented tasks and molecular-guided queries for molecular endpoint prediction, to retrieve supporting evidence balls and produce class-conditioned evidence representations for direct class-wise prediction. By introducing structured evidence units and task-relevant semantic guidance, EviBall reduces the reliance on learning an unconstrained global aggregation mechanism from scarce slide-level labels. It therefore reformulates few-shot WSI classification as structured evidence retrieval and competition among candidate classes. Extensive experiments across four morphology-oriented and molecular endpoint WSI tasks demonstrate that EviBall consistently outperforms conventional and vision-language MIL baselines under diverse few-shot settings, while providing spatially localized and class-specific evidence for each prediction.

cs.CV↗

arXiv2026

Bar-driven secular evolution largely complete in a disk galaxy 7.6 billion years ago

Disk galaxies like the Milky Way are thought to evolve through internal dynamical processes: the stellar disk forms a bar, the bar drives gas inflow that builds a nuclear stellar disk, and the bar vertically thickens into an X-shaped bulge. Although this evolution is thought to be slow, completing only at late cosmic times, its timing remains poorly constrained. We report James Webb Space Telescope imaging of a galaxy at redshift 0.92 (7.6 billion years ago) that already hosts an X-shaped bulge, a nuclear stellar disk, and an extended stellar disk, with geometry and inferred bar size indistinguishable from those of present-day barred galaxies. The X-shaped bulge marks the completion of the major phase of bar-driven evolution when the Universe was less than half its current age.

astro-ph.GA↗

arXiv2026

Looped World Models

Current world models face a fundamental tension: faithful long-horizon simulation demands deep computation, but deeper models are expensive to deploy and prone to compounding errors. We resolve this by introducing Looped World Models (LoopWM), which are the first looped architectures for world modelling. Our method iteratively refines latent environment states through a parameter-shared transformer block. This yield up to 100x parameter efficiency over conventional approaches with adaptive computation that automatically scales depth to match the complexity of each prediction step. Orthogonal to scaling model size and training data, LoopWM establishes iterative latent depth as a new scaling axis for world simulation, which might significantly push the community forward.

cs.LG↗

arXiv2026

AGE-MIL: Anchor-Guided Evidence Learning for Patient-Level Prediction

Existing computational pathology methods predominantly operate within whole-slide image (WSI)-level multiple instance learning (MIL) paradigms, while patient-level modeling remains underexplored. In routine pathological practice, however, pathologists derive diagnostic and prognostic conclusions by integrating evidence across multiple WSIs rather than relying on any single slide. This discrepancy creates a fundamental misalignment when patient-level supervision is directly imposed on conventional MIL frameworks, often leading to unstable optimization and degraded predictive reliability. To address this issue, we propose Anchor-Guided Evidence MIL (AGE-MIL), a weakly supervised framework for patient-level prediction. AGE-MIL constructs a patient-level anchor from slide representations to capture global pathological context and guide the retrieval and integration of diagnostically relevant local patches, enabling robust patient-level modeling. Patient-level risk is further modeled as an evidence accumulation process, promoting stable optimization under weak supervision. AGE-MIL is evaluated on six clinically relevant patient-level prediction tasks from two independent cohorts. Experimental results show that the proposed framework consistently outperforms eight state-of-the-art MIL methods. Code is available at https://github.com/wodeniua/AGE-MIL.

cs.CV↗

arXiv2026

DAG-Plan: Generating Directed Acyclic Dependency Graphs for Dual-Arm Cooperative Planning

Dual-arm robots promise greater efficiency but require planning for complex tasks with nonlinear sub-task dependencies. Current methods using Large Language Models (LLMs) suffer from a fundamental trade-off: generating linear sequences is efficient but fails to model parallelism and adapt to changes, while iterative querying is adaptive but too slow and costly. To bridge this gap, we introduce DAG-Plan, a novel task planning framework that for the first time employs a Directed Acyclic Graph (DAG) as the central representation for dual-arm coordination. The key insight is that a DAG natively captures complex sub-task dependencies and explicitly reveals opportunities for parallel execution. Within this framework, an LLM is used only once as a powerful semantic parser to translate a natural language instruction into a structured DAG. During execution, our system dynamically assigns candidate nodes to the suitable arm based on real-time environmental observations, enabling truly adaptive and parallel operation. Extensive evaluation on a dual-arm kitchen benchmark shows that DAG-Plan's structured approach fundamentally outperforms existing paradigms. It achieves a 48% higher success rate than single-query linear sequence methods with dual arm by robustly managing dependencies, and an 84.1% higher execution efficiency than iterative querying methods by eliminating the latency of repeated LLM calls. Our work demonstrates that a principled, graph-based representation is the key to unlocking efficient and reliable LLM-based planning for complex robotic systems. More demos and code are available on https://sites.google.com/view/dag-plan.

cs.RO↗

arXiv2026

Nanoscopic Multiplexing Optical Data Storage via Chip Fabrication

The accelerating growth of global data generation demands data storage platforms that offer high capacity, long lifespan, and low energy consumption beyond the limits of electronic memory technologies. Optical storage provides an attractive alternative. However, its density is fundamentally constrained by the optical diffraction limit and the limited scalability from the point-by-point laser writing, as well as thermal accumulation during high-speed writing. Here, we introduce a large-scale optical data storage scheme that is compatible with the progress in chip fabrication by combining electron-beam lithography (EBL) and ion implantation to deterministically encode high-density data. The approach achieves precise control of ion number and spatial distribution, enabling multi-bit grayscale encoding and wavelength division multiplexing with chip-scale patterning over millimeter areas. Wavelength-selective readout is performed using downconversion and upconversion fluorescence detection, allowing crosstalk-free retrieval of multiplexed data channels. We further develop a neural network-based super-resolution algorithm that reconstructs data beyond the diffraction limit, further increasing the effective storage density. Using this integrated framework, we achieve an optical data density of 10 Gbit/cm$^2$ with high fidelity. Our results establish a micro/nano-fabrication-compatible route to large-scale, high-density optical memory and provide a foundation for next-generation cold data optical storage technologies.

physics.optics↗

arXiv2026

PathNavigate: A Training-Free Pathology Agent with Surprise-Guided Scan and Shared Slide Memory for Whole-Slide Image VQA

Whole-slide image visual question answering (WSI-VQA) frames pathology as an extreme-context search problem: to answer a free-form clinical query, a system must first navigate a gigapixel slide under a strict inspection budget to locate sparse, high-resolution evidence. Existing approaches largely fall into two paradigms: i) supervised pathology multimodal large language models (MLLMs) and agents can absorb localization and reasoning into learned modules, but they often couple navigation to task-specific supervision and retraining, limiting their practicality; ii) training-free pathology agents avoid this cost by keeping core models frozen, but often follow a question-first design, constructing the initial candidate set mainly from query-conditioned relevance. This can miss decisive morphology that is not named in the question, and force heavier inference-time scaffolding. To address this challenge, we introduce PathNavigate, a training-free pathology agent built around a scan-search-readout routine. Before question matching, PathNavigate scans the current slide at low magnification with a shared online memory module over frozen pathology features, producing a slide-specific surprise field that marks an abnormal-region pool. It then applies question-conditioned PLIP relevance only within this pool to select high-magnification search targets. Finally, it extracts local high-magnification evidence and answers with a frozen perceptor-adjudicator stack, using the same online memory as slide-level context. Experiments on WSI-VQA and SlideBench-BCNB show that the proposed scan-search-readout design improves answer accuracy and yields more interpretable evidence-selection trajectories with higher efficiency.The code is available online.

cs.CV↗

arXiv2026

Vision Transformers for Preoperative CT-Based Prediction of Histopathologic Chemotherapy Response Score in High-Grade Serous Ovarian Carcinoma

Purpose. High-grade serous ovarian carcinoma (HGSOC) is characterized by pronounced biological and spatial heterogeneity and is frequently diagnosed at an advanced stage. Neoadjuvant chemotherapy (NACT) followed by delayed primary surgery is commonly employed in patients unsuitable for primary cytoreduction. The Chemotherapy Response Score (CRS) is a validated histopathological biomarker of response to NACT, but it is only available postoperatively. In this study, we investigate whether pre-treatment computed tomography (CT) imaging and clinical data can be used to predict CRS as an investigational decision-support adjunct to inform multidisciplinary team (MDT) discussions regarding expected treatment response. Methods. We proposed a 2.5D multimodal deep learning framework that processes lesion-dense omental slices using a pre-trained Vision Transformer encoder and integrates the resulting visual representations with clinical variables through an intermediate fusion module to predict CRS. Results. Our multimodal model, integrating imaging and clinical data, achieved a ROC-AUC of 0.95 alongside 95% accuracy and 80% precision on the internal test cohort (IEO, n=41 patients). On the external test set (OV04, n=70 patients), it achieved a ROC-AUC of 0.68, alongside 67% accuracy and 75% precision. Conclusion. These preliminary results demonstrate the feasibility of transformer-based deep learning for preoperative prediction of CRS in HGSOC using routine clinical data and CT imaging. As an investigational, pre-treatment decision-support tool, this approach may assist MDT discussions by providing early, non-invasive estimates of treatment response.

cs.CV↗

arXiv2026

CARE: A Molecular-Guided Foundation Model with Adaptive Region Modeling for Whole Slide Image Analysis

Foundation models have recently achieved impressive success in computational pathology, demonstrating strong generalization across diverse histopathology tasks. However, existing models overlook the heterogeneous and non-uniform organization of pathological regions of interest (ROIs) because they rely on natural image backbones not tailored for tissue morphology. Consequently, they often fail to capture the coherent tissue architecture beyond isolated patches, limiting interpretability and clinical relevance. To address these challenges, we present Cross-modal Adaptive Region Encoder (CARE), a foundation model for pathology that automatically partitions WSIs into several morphologically relevant regions. Specifically, CARE employs a two-stage pretraining strategy: (1) a self-supervised unimodal pretraining stage that learns morphological representations from 34,277 whole-slide images (WSIs) without segmentation annotations, and (2) a cross-modal alignment stage that leverages RNA and protein profiles to refine the construction and representation of adaptive regions. This molecular guidance enables CARE to identify biologically relevant patterns and generate irregular yet coherent tissue regions, selecting the most representative area as ROI. CARE supports a broad range of pathology-related tasks, using either the ROI feature or the slide-level feature obtained by aggregating adaptive regions. Based on only one-tenth of the pretraining data typically used by mainstream foundation models, CARE achieves superior average performance across 33 downstream benchmarks, including morphological classification, molecular prediction, and survival analysis, and outperforms other foundation model baselines overall.

cs.CV↗

arXiv2026

PCodeTrans: Translate Decompiled Pseudocode to Compilable and Executable Equivalent

Decompilation is foundational to binary analysis, yet conventional tools prioritize human readability over strict recompilability and verifiable runtime correctness. While recent LLM-based approaches attempt to refine decompiled pseudocode, they typically either optimize solely for readability or rely on static analysis for evaluation. This makes them prone to "semantic hallucinations" that compromise accuracy and fail to resolve actual runtime failures. For critical tasks like software modernization and vulnerability remediation, recovered code must not only compile but replicate the original binary's behavior. We present PCodeTrans, a feedback-driven framework that bridges the gap between decompilation, recompilation, and rigorous function-level dynamic validation. After extracting a minimal yet coherent context to guarantee recompilability, PCodeTrans employs an in situ substitutable engine to hot-swap the compiled function directly into the unmodified binary, natively preserving its authentic execution context and global dependencies. Guided by fine-grained differential tracing, PCodeTrans generates precise runtime feedback to iteratively guide an LLM in repairing semantic discrepancies. Evaluated on Coreutils and Binutils, PCodeTrans achieves unprecedented recovery performance when rectifying raw Hex-Rays outputs, attaining 100% function-level compilability on unstripped binaries alongside 99.55% and 99.89% test-validated behavioral consistency, respectively. In doing so, it resolves 76.56% and 79.74% of logic errors exposed by official test suites. Exhibiting exceptional resilience, PCodeTrans maintains over 96% behavioral consistency even on fully stripped binaries. By significantly outperforming all existing baselines, PCodeTrans paves a practical path to reliably translate decompiled pseudocode into compilable and executable equivalents.

cs.SE↗

arXiv2026

SecCodeBench-V2 Technical Report

We introduce SecCodeBench-V2, a publicly released benchmark for evaluating Large Language Model (LLM) copilots' capabilities of generating secure code. SecCodeBench-V2 comprises 98 generation and fix scenarios derived from Alibaba Group's industrial productions, where the underlying security issues span 22 common CWE (Common Weakness Enumeration) categories across five programming languages: Java, C, Python, Go, and JavaScript. SecCodeBench-V2 adopts a function-level task formulation: each scenario provides a complete project scaffold and requires the model to implement or patch a designated target function under fixed interfaces and dependencies. For each scenario, SecCodeBench-V2 provides executable proof-of-concept (PoC) test cases for both functional validation and security verification. All test cases are authored and double-reviewed by security experts, ensuring high fidelity, broad coverage, and reliable ground truth. Beyond the benchmark itself, we build a unified evaluation pipeline that assesses models primarily via dynamic execution. For most scenarios, we compile and run model-generated artifacts in isolated environments and execute PoC test cases to validate both functional correctness and security properties. For scenarios where security issues cannot be adjudicated with deterministic test cases, we additionally employ an LLM-as-a-judge oracle. To summarize performance across heterogeneous scenarios and difficulty levels, we design a Pass@K-based scoring protocol with principled aggregation over scenarios and severity, enabling holistic and comparable evaluation across models. Overall, SecCodeBench-V2 provides a rigorous and reproducible foundation for assessing the security posture of AI coding assistants, with results and artifacts released at https://alibaba.github.io/sec-code-bench. The benchmark is publicly available at https://github.com/alibaba/sec-code-bench.

cs.CR↗

arXiv2026

HAAF: Hierarchical Adaptation and Alignment of Foundation Models for Few-Shot Pathology Anomaly Detection

Precision pathology relies on detecting fine-grained morphological abnormalities within specific Regions of Interest (ROIs), as these local, texture-rich cues - rather than global slide contexts - drive expert diagnostic reasoning. While Vision-Language (V-L) models promise data efficiency by leveraging semantic priors, adapting them faces a critical Granularity Mismatch, where generic representations fail to resolve such subtle defects. Current adaptation methods often treat modalities as independent streams, failing to ground semantic prompts in ROI-specific visual contexts. To bridge this gap, we propose the Hierarchical Adaptation and Alignment Framework (HAAF). At its core is a novel Cross-Level Scaled Alignment (CLSA) mechanism that enforces a sequential calibration order: visual features first inject context into text prompts to generate content-adaptive descriptors, which then spatially guide the visual encoder to spotlight anomalies. Additionally, a dual-branch inference strategy integrates semantic scores with geometric prototypes to ensure stability in few-shot settings. Experiments on four benchmarks show HAAF significantly outperforms state-of-the-art methods and effectively scales with domain-specific backbones (e.g., CONCH) in low-resource scenarios.

cs.CV↗

arXiv2026

Information-Theoretic Generalization Bounds of Replay-based Continual Learning

Continual learning (CL) has emerged as a dominant paradigm for acquiring knowledge from sequential tasks while avoiding catastrophic forgetting. Although many CL methods have been proposed to show impressive empirical performance, the theoretical understanding of their generalization behavior remains limited, particularly for replay-based approaches. This paper establishes a unified theoretical framework for replay-based CL, deriving a series of information-theoretic generalization bounds that explicitly elucidate the impact of the memory buffer alongside the current task on generalization performance. Specifically, our hypothesis-based bounds capture the trade-off between the number of selected exemplars and the information dependency between the hypothesis and the memory buffer. Our prediction-based bounds yield tighter and computationally tractable upper bounds on the generalization error by leveraging low-dimensional variables. Theoretical analysis is general and broadly applicable to a wide range of learning algorithms, exemplified by stochastic gradient Langevin dynamics (SGLD) as a representative method. Comprehensive experimental evaluations demonstrate the effectiveness of our derived bounds in capturing the generalization dynamics in replay-based CL settings.

cs.LG↗