arXiv subjects

Pengju Wang

Publications and source records attributed to Pengju Wang.

At least 19 recordsLinked to original sources

arXiv2026

ClinFusion: A Vision-Centric Multimodal LLM System for Holistic Medical Understanding

Multimodal large language models (MLLMs) hold immense potential to revolutionize clinical practice, yet deploying them in the medical domain is fundamentally a vision-centric challenge: models must absorb knowledge from heterogeneous 2D and 3D medical images, and evaluation protocols must align with radiologists' clinical practice and provide an accurate, fine-grained and factualness-driven assessment. In this paper, we introduce ClinFusion, a vision-centric MLLM designed for holistic medical understanding that systematically addresses these limitations. We propose a compositional and cascaded vision encoder architecture featuring a Cascade Spatial-Aware Locality Fusion operator that unifies diverse 2D and native 3D medical image understanding within a fused encoder. We further introduce a vision-grounded evaluation framework, including MedIF-Bench for instruction-following assessment and a region-of-interest-grounded method for clinically aligned and factualness-driven report generation evaluation. We show that ClinFusion sets a new state-of-the-art across a comprehensive suite of 2D and 3D multimodal medical benchmarks---spanning visual question answering, report generation, and instruction following---as well as textual medical tasks, outperforming leading open-source medical MLLMs (\textit{e.g.}, Hulu-Med, Lingshu) on 20 out of 24 benchmarks and demonstrating multimodal capabilities better than powerful proprietary models such as GPT-5.2 and Gemini-3-Flash on 13 out of 16 benchmarks, and can be further augmented with agentic tool use for retrieval-augmented and tool-assisted clinical workflows. A blinded evaluation by board-certified radiologists confirms that ClinFusion produces the highest-ranked reports, and validates our RoI-grounded metric as achieving the strongest correlation with expert judgment among all automatic evaluation metrics examined.

cs.CV↗

arXiv2026

Rapid quasi-periodic reconfiguration of the accretion column in pulsar 1A 0535+262

Accretion onto strongly magnetized neutron stars is commonly interpreted using quasi-steady models, in which the accretion-column structure adjusts smoothly to the mass inflow rate. The cyclotron line in the X-ray spectrum, whose centroid energy traces the magnetic field strength and thus the height of the line-forming region, provides a key diagnostic of this structure. Whether this simple quasi-steady description remains valid on short dynamical timescales has remained uncertain. Here we show that, during a giant outburst of the X-ray pulsar 1A~0535+262, quasi-periodic hard X-ray flux variations are accompanied by synchronized oscillations of the cyclotron line energy, with amplitudes exceeding those expected from simple accretion-rate fluctuations. The anti-correlation between cyclotron energy and apparent flux provides direct spectral-timing evidence for rapid changes in the line-forming region, which we interpret as geometric reconfiguration of the accretion column. The variability emerges in the luminosity regime where radiation pressure becomes dynamically important. These results reveal limitations of a simple quasi-steady interpretation for this source and suggest that radiation-supported columns can enter intrinsically dynamical states in high-luminosity accreting pulsars.

astro-ph.HE↗

arXiv2026

Agentic Fusion of Large Atomic and Language Models to Accelerate Superconductor Discovery

Artificial intelligence has accelerated materials discovery through high-throughput prediction and generation, yet the decision problem remains a formidable bottleneck. While current AI systems readily propose millions of candidates, navigating the decision regarding a viable experimental target requires resolving multi-dimensional judgments across atomic-scale numerical computation and high-level semantic reasoning. Here we present ElementsClaw, an agentic framework for materials discovery that orchestrates a suite of Large Atomic Model (LAM) tools finetuned from our proposed 1-billion-parameter model Elements for numerical computation, while leveraging Large Language Models (LLMs) for semantic reasoning. Applied to superconductors, ElementsClaw rediscovers 66 experimentally verified superconductors that are absent from the standard SuperCon3D database. Scaling to 2.4 million equilibrium crystals, ElementsClaw identifies 68,000 high-confidence candidates in just 28 GPU hours (https://developer.damo-academy.com/material), expanding known superconducting space by orders of magnitude compared to datasets curated over decades. Guided by the agent's reasoning, we experimentally synthesize and verify four novel superconductors: the motif-guided Zr$_3$ScRe$_8$ ($T_c$ = 6.5 K), the de novo generated HfZrRe$_4$ ($T_c$ = 5.9 K), the structurally reinterpreted Zr$_4$VRe$_7$ ($T_c$ = 3.5 K), and the database-latent Hf$_{21}$Re$_{25}$ ($T_c$ = 2.5 K). Together, our results establish a knowledge integrated, autonomously orchestrated, and experimentally grounded paradigm for materials discovery.

cs.LG↗

arXiv2026

Privacy-Preserving Model Transcription with Differentially Private Synthetic Distillation

While many deep learning models trained on private datasets have been deployed in various practical tasks, they may pose a privacy leakage risk as attackers could recover informative data or label knowledge from models. In this work, we present \emph{privacy-preserving model transcription}, a data-free model-to-model conversion solution to facilitate model deployment with a privacy guarantee. To this end, we propose a cooperative-competitive learning approach termed \emph{differentially private synthetic distillation} that learns to convert a pretrained model (teacher) into its privacy-preserving counterpart (student) via a trainable generator without access to private data. The learning collaborates with three players in a unified framework and performs alternate optimization: i)~the generator is learned to generate synthetic data, ii)~the teacher and student accept the synthetic data and compute differential private labels by flexible data or label noisy perturbation, and iii)~the student is updated with noisy labels and the generator is updated by taking the student as a discriminator for adversarial training. We theoretically prove that our approach can guarantee differential privacy and convergence. The transcribed student has good performance and privacy protection, while the resulting generator can generate private synthetic data for downstream tasks. Extensive experiments clearly demonstrate that our approach outperforms 26 state-of-the-arts.

cs.LG↗

arXiv2025

Observatory Science with eXTP

Scheduled for launch in 2030, the enhanced X-ray Timing and Polarization (eXTP) telescope is a Chinese space-based mission aimed at studying extreme conditions and phenomena in astrophysics. eXTP will feature three main payloads: Spectroscopy Focusing Arrays (SFAs), Polarimetry Focusing Arrays (PFAs), and a Wide-field Camera (W2C). This white paper outlines observatory science, incorporating key scientific advances and instrumental changes since the publication of the previous white paper [1]. We will discuss perspectives of eXTP on the research domains of flare stars, supernova remnants, pulsar wind nebulae, cataclysmic variables, X-ray binaries, ultraluminous X-ray sources, AGN, and pulsar-based positioning and timekeeping.

astro-ph.IM↗

arXiv2025

Probing the emission geometry of the X-ray pulsar 2S 1417$-$624 during a weak outburst with NICER, IXPE, and NuSTAR

We report results from a multi-mission observational campaign of the transient X-ray pulsar 2S~1417$-$624 during its 2025 outburst, using data from NICER, IXPE, and NuSTAR. Phase-averaged and phase-resolved spectroscopy with NICER and NuSTAR reveal that a typical broken power-law model with a high-energy cut-off well describes the broadband spectra. Several spectral parameters, however, show clear and systematic modulations with pulse phase, indicating variations in the physical conditions of the emitting plasma over the neutron star's rotation. IXPE provides the first polarimetric measurements of this source, yielding a phase-averaged polarization degree (PD) of $4.8 \pm 1.2$% and a polarization angle (PA) of ${17}^{\circ} \pm {7}^{\circ}$, both quoted at the $1\sigma$ confidence level. Fitting the phase-resolved PA with the rotating vector model (RVM) gives a magnetic obliquity of $\theta = 69_{-29}^{+13}$ deg, indicating a significantly inclined magnetic geometry that may approach a quasi-orthogonal configuration. In addition, using the unbinned photon-by-photon method, we obtain a PD of $5.9 \pm 1.2$% across the pulse phase, together with a pulsar geometry consistent with that inferred from the binned analysis, assuming the variable PA predicted by the RVM. A simultaneous RVM fit across the three energy bands, 2--5 keV, 5--6 keV, and 6--8 keV, provides the strongest constraints on the geometrical parameters, yielding $\theta = {84}_{-6}^{+4}$ deg. Together, these findings demonstrate pronounced phase-dependent spectral and polarization variability, offering valuable constraints on the geometry and emission processes within the accretion region of this transient X-ray pulsar.

astro-ph.HE↗

arXiv2025

Probing the polarized emission from the accretion-powered pulsar 4U 1907+09 with IXPE

We present observations of the accretion-powered X-ray pulsar (XRP) 4U 1907+09, conducted with the Imaging X-ray Polarimetry Explorer (IXPE) that delivers the first high-quality polarization measurements of this source. 4U 1907+09 was observed twice during its brightest periods close to the periastron. We observe a stronger polarization in the first observation, with the phase-averaged polarization degree (PD) of $6.0 \pm 1.6\%$ and polarization angle (PA) of $69^\circ \pm 8^\circ$. In contrast, the second observation provides weaker constraints on the polarimetric properties, with a PD=$2.2 \pm 1.6\%$ and a PA=$46^\circ \pm 23^\circ$, as determined from the spectral-polarimetric analysis. Combining the data from both observations results in a PD=$3.7 \pm 1.1\%$ and a PA=$63^\circ \pm 9^\circ$. We detect an energy-dependent PA in the phase-averaged analyses with a significance of 1.7 $σ$. In the phase-resolved analyses, we observe a potential PA rotation by approximately $90^\circ$ between adjacent energy bands (4--5 and 5--6 keV) within the single phase bin of 0.25--0.375. We also investigate the influence of short flares on the polarization properties of this source. The results suggest that flares do not significantly affect the energy-phase-dependent PA, implying that the pulsar's geometry remains stable during flare events.

astro-ph.HE↗

arXiv2025

DiffSpectra: Molecular Structure Elucidation from Spectra using Diffusion Models

Molecular structure elucidation from spectra is a fundamental challenge in molecular science. Conventional approaches rely heavily on expert interpretation and lack scalability, while retrieval-based machine learning approaches remain constrained by limited reference libraries. Generative models offer a promising alternative, yet most adopt autoregressive architectures that overlook 3D geometry and struggle to integrate diverse spectral modalities. In this work, we present DiffSpectra, a generative framework that formulates molecular structure elucidation as a conditional generation process, directly inferring 2D and 3D molecular structures from multi-modal spectra using diffusion models. Its denoising network is parameterized by the Diffusion Molecule Transformer, an SE(3)-equivariant architecture for geometric modeling, conditioned by SpecFormer, a Transformer-based spectral encoder capturing multi-modal spectral dependencies. Extensive experiments demonstrate that DiffSpectra accurately elucidates molecular structures, achieving 40.76% top-1 and 99.49% top-10 accuracy. Its performance benefits substantially from 3D geometric modeling, SpecFormer pre-training, and multi-modal conditioning. To our knowledge, DiffSpectra is the first framework that unifies multi-modal spectral reasoning and joint 2D/3D generative modeling for de novo molecular structure elucidation.

cs.LG↗

arXiv2024

Towards Personalized Federated Learning via Comprehensive Knowledge Distillation

Federated learning is a distributed machine learning paradigm designed to protect data privacy. However, data heterogeneity across various clients results in catastrophic forgetting, where the model rapidly forgets previous knowledge while acquiring new knowledge. To address this challenge, personalized federated learning has emerged to customize a personalized model for each client. However, the inherent limitation of this mechanism is its excessive focus on personalization, potentially hindering the generalization of those models. In this paper, we present a novel personalized federated learning method that uses global and historical models as teachers and the local model as the student to facilitate comprehensive knowledge distillation. The historical model represents the local model from the last round of client training, containing historical personalized knowledge, while the global model represents the aggregated model from the last round of server aggregation, containing global generalized knowledge. By applying knowledge distillation, we effectively transfer global generalized knowledge and historical personalized knowledge to the local model, thus mitigating catastrophic forgetting and enhancing the general performance of personalized models. Extensive experimental results demonstrate the significant advantages of our method.

cs.LG↗

arXiv2024

Personalized Federated Learning via Backbone Self-Distillation

In practical scenarios, federated learning frequently necessitates training personalized models for each client using heterogeneous data. This paper proposes a backbone self-distillation approach to facilitate personalized federated learning. In this approach, each client trains its local model and only sends the backbone weights to the server. These weights are then aggregated to create a global backbone, which is returned to each client for updating. However, the client's local backbone lacks personalization because of the common representation. To solve this problem, each client further performs backbone self-distillation by using the global backbone as a teacher and transferring knowledge to update the local backbone. This process involves learning two components: the shared backbone for common representation and the private head for local personalization, which enables effective global knowledge transfer. Extensive experiments and comparisons with 12 state-of-the-art approaches demonstrate the effectiveness of our approach.

cs.LG↗

arXiv2024

Federated Learning with Label-Masking Distillation

Federated learning provides a privacy-preserving manner to collaboratively train models on data distributed over multiple local clients via the coordination of a global server. In this paper, we focus on label distribution skew in federated learning, where due to the different user behavior of the client, label distributions between different clients are significantly different. When faced with such cases, most existing methods will lead to a suboptimal optimization due to the inadequate utilization of label distribution information in clients. Inspired by this, we propose a label-masking distillation approach termed FedLMD to facilitate federated learning via perceiving the various label distributions of each client. We classify the labels into majority and minority labels based on the number of examples per class during training. The client model learns the knowledge of majority labels from local data. The process of distillation masks out the predictions of majority labels from the global model, so that it can focus more on preserving the minority label knowledge of the client. A series of experiments show that the proposed approach can achieve state-of-the-art performance in various cases. Moreover, considering the limited resources of the clients, we propose a variant FedLMD-Tf that does not require an additional teacher, which outperforms previous lightweight approaches without increasing computational costs. Our code is available at https://github.com/wnma3mz/FedLMD.

cs.LG↗

arXiv2024

Privacy-Preserving Student Learning with Differentially Private Data-Free Distillation

Deep learning models can achieve high inference accuracy by extracting rich knowledge from massive well-annotated data, but may pose the risk of data privacy leakage in practical deployment. In this paper, we present an effective teacher-student learning approach to train privacy-preserving deep learning models via differentially private data-free distillation. The main idea is generating synthetic data to learn a student that can mimic the ability of a teacher well-trained on private data. In the approach, a generator is first pretrained in a data-free manner by incorporating the teacher as a fixed discriminator. With the generator, massive synthetic data can be generated for model training without exposing data privacy. Then, the synthetic data is fed into the teacher to generate private labels. Towards this end, we propose a label differential privacy algorithm termed selective randomized response to protect the label information. Finally, a student is trained on the synthetic data with the supervision of private labels. In this way, both data privacy and label privacy are well protected in a unified framework, leading to privacy-preserving models. Extensive experiments and analysis clearly demonstrate the effectiveness of our approach.

cs.LG↗

arXiv2024

Learning Privacy-Preserving Student Networks via Discriminative-Generative Distillation

While deep models have proved successful in learning rich knowledge from massive well-annotated data, they may pose a privacy leakage risk in practical deployment. It is necessary to find an effective trade-off between high utility and strong privacy. In this work, we propose a discriminative-generative distillation approach to learn privacy-preserving deep models. Our key idea is taking models as bridge to distill knowledge from private data and then transfer it to learn a student network via two streams. First, discriminative stream trains a baseline classifier on private data and an ensemble of teachers on multiple disjoint private subsets, respectively. Then, generative stream takes the classifier as a fixed discriminator and trains a generator in a data-free manner. After that, the generator is used to generate massive synthetic data which are further applied to train a variational autoencoder (VAE). Among these synthetic data, a few of them are fed into the teacher ensemble to query labels via differentially private aggregation, while most of them are embedded to the trained VAE for reconstructing synthetic data. Finally, a semi-supervised student learning is performed to simultaneously handle two tasks: knowledge transfer from the teachers with distillation on few privately labeled synthetic data, and knowledge enhancement with tangent-normal adversarial regularization on many triples of reconstructed synthetic data. In this way, our approach can control query cost over private data and mitigate accuracy degradation in a unified manner, leading to a privacy-preserving student model. Extensive experiments and analysis clearly show the effectiveness of the proposed approach.

cs.LG↗

arXiv2024

Private Gradient Estimation is Useful for Generative Modeling

While generative models have proved successful in many domains, they may pose a privacy leakage risk in practical deployment. To address this issue, differentially private generative model learning has emerged as a solution to train private generative models for different downstream tasks. However, existing private generative modeling approaches face significant challenges in generating high-dimensional data due to the inherent complexity involved in modeling such data. In this work, we present a new private generative modeling approach where samples are generated via Hamiltonian dynamics with gradients of the private dataset estimated by a well-trained network. In the approach, we achieve differential privacy by perturbing the projection vectors in the estimation of gradients with sliced score matching. In addition, we enhance the reconstruction ability of the model by incorporating a residual enhancement module during the score matching. For sampling, we perform Hamiltonian dynamics with gradients estimated by the well-trained network, allowing the sampled data close to the private dataset's manifold step by step. In this way, our model is able to generate data with a resolution of 256x256. Extensive experiments and analysis clearly demonstrate the effectiveness and rationality of the proposed approach.

cs.CV↗

arXiv2024

Learning Differentially Private Diffusion Models via Stochastic Adversarial Distillation

While the success of deep learning relies on large amounts of training datasets, data is often limited in privacy-sensitive domains. To address this challenge, generative model learning with differential privacy has emerged as a solution to train private generative models for desensitized data generation. However, the quality of the images generated by existing methods is limited due to the complexity of modeling data distribution. We build on the success of diffusion models and introduce DP-SAD, which trains a private diffusion model by a stochastic adversarial distillation method. Specifically, we first train a diffusion model as a teacher and then train a student by distillation, in which we achieve differential privacy by adding noise to the gradients from other models to the student. For better generation quality, we introduce a discriminator to distinguish whether an image is from the teacher or the student, which forms the adversarial training. Extensive experiments and analysis clearly demonstrate the effectiveness of our proposed method.

cs.LG↗

arXiv2024

Evolution of QPOs in GX 339-4 and EXO 1846-031 with Insight-HXMT and NICER

We conduct a spectral and timing analysis of GX 339-4 and EXO 1846-031 with the aim of studying the evolution of Type-C QPOs with spectral parameters. The high cadence data from Insight-HXMT and NICER allow us to track them. Type-C QPOs appear at the end of low-hard state and/or hard-intermediate state. The results reveal that the QPO frequency is closely related to the inner disk radius and mass accretion rate in the two sources. Such a correlation is nicely consistent with the dynamic frequency model.

astro-ph.HE↗

arXiv2024

Recovery of High-energy Low-frequency Quasi-periodic Oscillations from Black Hole X-ray Binary MAXI J1535-571 with a Hilbert-Huang Transform Method

We propose a method based on the Hilbert-Huang transform (HHT) to recover the high-energy waveform of low-frequency quasi-periodic oscillations (LFQPOs). Based on the method, we successfully obtain the modulation of the phase-folded light curve above 170 keV using the QPO phase reconstructed at lower energies in MAXI J1535-571 with Insight-HXMT observations. A comprehensive simulation study is conducted to demonstrate that such modulation indeed originates from the QPO. Thus the highest energies turn out to significantly exceed the upper limit of ~100 keV for QPOs reported previously using the Fourier method, marking the first opportunity to study QPO properties above 100 keV in this source. Detailed analyses of these high-energy QPO profiles reveal different QPO properties between the 30-100 keV and 100-200 keV energy ranges: the phase lag remains relatively stable, and the amplitude slightly increases below ~100 keV, whereas above this threshold, soft phase lags and a decrease in amplitude are observed. Given the reports of a hard tail detection in broad spectroscopy, we propose that the newly discovered QPO properties above 100 keV are dominated by the hard tail component, possibly stemming from a relativistic jet. Our findings also indicate a strong correlation between the QPOs originating from the jet and corona, supporting the scenario of jet-corona coupling precssion. We emphasize that our proposed HHT-based method can serve as an efficient manner in expanding the high energy band for studying QPOs, thereby enhancing our understanding of their origin.

astro-ph.HE↗

arXiv2024

Cyclotron line evolution revealed with pulse-to-pulse analysis in the 2020 outburst of 1A 0535+262

We present a detailed analysis of the X-ray luminosity (Lx) dependence of the cyclotron absorption line energy (Ecyc) for the X-ray binary pulsar 1A 0535+262 during its 2020 giant outburst based on pulse-to-pulse analysis. By applying this technique to high cadence observations of Insight-HXMT, we reveal the most comprehensive Ecyc-Lx correlation across a broad luminosity range of ~(0.03-1.3)*10^38 erg/s. Apart from the positive and negative correlations between cyclotron line energy and luminosity at Lx~(1-3)*10^37 erg/s and ~(7-13)*10^37 erg/s, which are expected from the typical subcritical and supercritical accretion regimes, respectively, a plateau in the correlation is also detected at ~(3-7)*10^37 erg/s^-1. Moreover, at the lowest luminosity level (Lx<10^37 erg/s), the positive Ecyc-Lx correlation seems to be broken, and the pulse profile also occurs a significant transition. These discoveries provide the first complete view on the correlation between luminosity and the centriod energy of the cyclotron line, and therefore are relevant for understanding how accretion onto magnetized neutron stars depends on luminosity.

astro-ph.HE↗