arXiv subjects

Hu Chen

Publications and source records attributed to Hu Chen.

At least 19 recordsLinked to original sources

arXiv2026

Hallucination Detection and Correction in Medical VLMs via Counter-Evidence Verification

Vision-Language models (VLMs) reliability in medical diagnosis is challenged by trust-undermining hallucinations. Existing hallucination detection approaches mainly focus on identifying factual inconsistencies between generated text and reference data. While some studies analyze where models attend in images, they seldom verify whether such attention truly reflects the visual evidence supporting the generated text. To address this gap, we propose Co}unter-Evidence Verification (CoEV), a training-free plug-and-play framework that detects and corrects hallucinations through evidence-based factual consistency verification. CoEV performs bidirectional verification between textual assertions and visual evidence, testing whether each statement is supported by its corresponding evidence region, and assigns each statement into a four-quadrant diagnostic map capturing combinations of text factuality and visual grounding. CoEV detects hallucinated content and serves as a post hoc refinement tool, correcting hallucinations without retraining. Extensive experiments on four medical datasets show that CoEV combats hallucinations in VLMs.For hallucination detection, CoEV consistently outperforms existing methods, improving average PR-AUC and ROC-AUC by 3.0% and 3.9% absolute points respectively, with notable gains of up to 18.5% in specific VQA scenarios. For hallucination correction, it improves Micro-F1 by up to 12.5%, reduces hallucination rates by over 11.9% on medical report generation, and also boosts medical VQA accuracy. These results show that CoEV enables reliable detection and correction of hallucinations, providing clinicians with dependable, evidence-based cues for diagnosis. Code will be released upon acceptance.

cs.CV↗

arXiv2026

PULSE: Generative Phase Evolution for Non-Stationary Time Series Forecasting

Time series forecasting under non-stationarity faces a fundamental tension between capturing stable representations and adapting to distribution shifts. Existing methods implicitly rely on static historical assumptions, leading to a critical failure mode we term Phase Amnesia, where models become blind to the evolving global context. To resolve this, we formalize non-stationary dynamics through three physical hypotheses: wold decomposition, dynamical phase evolution, and heteroscedastic manifold generation. These principles inspire PULSE, a physics-informed, plug-and-play framework adopting a Disentangle--Evolve--Simulate design philosophy. Specifically, PULSE utilizes phase-anchored disentanglement to resolve optimization interference caused by dominant trends, employs a Phase Router to actively generate future trajectories, and introduces Statistic-Aware Mixup (SAM) to ensure robustness against out-of-distribution volatility. Empirically, PULSE enables a simple MLP backbone to achieve state-of-the-art or highly competitive performance across 12 real-world benchmarks. This validates that a correct physics-informed inductive bias is far more critical than raw architectural complexity for non-stationary forecasting. The code is available at: https://github.com/Gemost/PULSE.

cs.LG↗

arXiv2026

WaveSFNet: A Wavelet-Based Codec and Spatial--Frequency Dual-Domain Gating Network for Spatiotemporal Prediction

Spatiotemporal predictive learning aims to forecast future frames from historical observations in an unsupervised manner, and is critical to a wide range of applications. The key challenge is to model long-range dynamics while preserving high-frequency details for sharp multi-step predictions. Existing efficient recurrent-free frameworks typically rely on strided convolutions or pooling for sampling, which tends to discard textures and boundaries, while purely spatial operators often struggle to balance local interactions with global propagation. To address these issues, we propose WaveSFNet, an efficient framework that unifies a wavelet-based codec with a spatial--frequency dual-domain gated spatiotemporal translator. The wavelet-based codec preserves high-frequency subband cues during downsampling and reconstruction. Meanwhile, the translator first injects adjacent-frame differences to explicitly enhance dynamic information, and then performs dual-domain gated fusion between large-kernel spatial local modeling and frequency-domain global modulation, together with gated channel interaction for cross-channel feature exchange. Extensive experiments demonstrate that WaveSFNet achieves competitive prediction accuracy on Moving MNIST, TaxiBJ, and WeatherBench, while maintaining low computational complexity. Our code is available at https://github.com/fhjdqaq/WaveSFNet.

cs.CV↗

arXiv2026

RareCollab: an LLM-powered framework for multimodal reasoning in Mendelian disease diagnosis

Rare disease diagnosis increasingly relies on integrating genomic, phenotypic and transcriptomic evidence, yet these signals remain difficult to reconcile within a common interpretive framework. Here we present RareCollab, an LLM-powered framework for multimodal reasoning in Mendelian disease diagnosis that integrates more than 100 diagnostic evidence signals across DNA, RNA, phenotype, curated variant-level knowledge, and in-silico pathogenicity evidence. This design enables large language models to operate as calibrated, interpretable reasoning modules rather than as a single end-to-end ranker. We applied RareCollab to 890 patients from three cohorts, including 119 Undiagnosed Diseases Network probands with paired DNA and RNA data, constituting a large systematic benchmark for multimodal rare disease diagnosis under paired genomic and transcriptomic evaluation. In this real-world multimodal benchmark, RareCollab prioritized 94% of diagnostic genes within the top 10. Across recall thresholds from top 1 to top 10, it consistently outperformed proprietary phenotype-driven LLM baselines including Claude Sonnet 4.6 and GPT-5-mini by more than 25% on average and surpassed established state-of-the-art variant prioritization methods by 11%-24%. RareCollab also reshapes the diagnostic contribution of RNA evidence, which contributes to prioritization of the diagnostic gene in 35% of cases (42/119). Together, these results establish RareCollab as a scalable and interpretable framework for multimodal rare disease diagnosis.

q-bio.GN↗

arXiv2026

Local convergence analysis of a linearized Alikhanov scheme for the time fractional sine-Gordon equation

This paper investigates the time fractional sine-Gordon equation whose solution exhibits a weak singularity of type t^{\alpha}. By means of the Alikhanov formula we derive a fully discrete, linearized scheme. Using the more general regularity assumption, we derive a sharp truncation-error bound for the fractional derivative. Furthermore, we prove a key inequality and a less restrictive stability result that is valid on general graded temporal meshes. Consequently, the temporal local convergence order is shown to be min{2, r} in H^1-seminorm, where r is the degree of grading; numerical experiments confirm that the optimal rate is already attained as soon as r = 2.

math.NA↗

arXiv2026

Pointwise-in-time convergence analysis of an Alikhanov scheme for a 2D nonlinear subdiffusion equation

In this paper, we discretize the Caputo time derivative of order \alpha \in (0,1) using the Alikhanov scheme on a quasi-graded temporal mesh, and employ the Newton linearization method to approximate the nonlinear term. This yields a linearized fully discrete scheme for the two-dimensional nonlinear time fractional subdiffusion equation with weakly singular solutions. For the purpose of conducting a pointwise convergence analysis using the comparison principle, we develop a new stability result. The global L^2-norm convergence order is min{\alpha r, 2}, and the local L^2-norm convergence order is min{r, 2} under appropriate conditions and assumptions. Ultimately, the rates of convergence demonstrated by the numerical experiments serve to validate the analytical outcomes.

math.NA↗

arXiv2025

Intelligent Communication Mixture-of-Experts Boosted-Medical Image Segmentation Foundation Model

Foundation models for medical image segmentation have achieved remarkable performance. Adaptive fine-tuning of natural image segmentation foundation models is crucial for medical image segmentation tasks. However, some limitations exist in existing fine-tuning methods: 1) insufficient representation of high-level features and 2) the fine-tuning process disrupts the structural integrity of pretrained weights. Inspired by these critical problems, we propose an intelligent communication mixture-of-experts boosted-medical image segmentation foundation model, named IC-MoE, with twofold ideas: 1) We construct basic experts, semantic experts, and adaptive experts. Moreover, we implement a pixel probability adaptive voting strategy, which enables expert selection and fusion through label consistency and load balancing. This approach preliminarily enhances the representation capability of high-level features while preserving the structural integrity of pretrained weights. 2) We propose a semantic-guided contrastive learning method to address the issue of weak supervision in contrastive learning. This method further enhances the representation capability of high-level features while preserving the structural integrity of pretrained weights. Extensive experiments across three public medical image segmentation datasets demonstrate that the IC-MoE outperforms other SOTA models. Consequently, the proposed IC-MoE effectively supplements foundational medical image segmentation models with high-level features and pretrained structural integrity. We also validate the superior generalizability of the IC-MoE across diverse medical image segmentation scenarios.

cs.CV↗

arXiv2025

Dual-Forward Path Teacher Knowledge Distillation: Bridging the Capacity Gap Between Teacher and Student

Knowledge distillation (KD) provides an effective way to improve the performance of a student network under the guidance of pre-trained teachers. However, this approach usually brings in a large capacity gap between teacher and student networks, limiting the distillation gains. Previous methods addressing this problem either discard accurate knowledge representation or fail to dynamically adjust the transferred knowledge, which is less effective in addressing the capacity gap problem and hinders students from achieving comparable performance with the pre-trained teacher. In this work, we extend the ideology of prompt-based learning to address the capacity gap problem, and propose Dual-Forward Path Teacher Knowledge Distillation (DFPT-KD), which replaces the pre-trained teacher with a novel dual-forward path teacher to supervise the learning of student. The key to DFPT-KD is prompt-based tuning, i.e., establishing an additional prompt-based forward path within the pre-trained teacher and optimizing it with the pre-trained teacher frozen to make the transferred knowledge compatible with the representation ability of the student. Extensive experiments demonstrate that DFPT-KD leads to trained students performing better than the vanilla KD. To make the transferred knowledge better compatible with the representation abilities of the student, we further fine-tune the whole prompt-based forward path, yielding a novel distillation approach dubbed DFPT-KD+. By extensive experiments, it is shown that DFPT-KD+ improves upon DFPT-KD and achieves state-of-the-art accuracy performance.

cs.LG↗

arXiv2025

Photonic Energy-Coherence Theorem and Experimental Validations

Wave-particle duality, intertwining two inherently contradictory properties of quantum systems, remains one of the most conceptually profound aspects of quantum mechanics. By using the concept of energy capacity, the ability of a quantum system to store and extract energy, we derive a device-independent uncertainty relation for wave-particle duality. This relation is shown to be independent of both the representation space and the measurement basis of the quantum system. Furthermore, we experimentally validate this wave-particle duality relation using a photon-based platform.

quant-ph↗

arXiv2025

An efffcient numerical scheme for two-dimensional nonlinear time fractional Schr\"odinger equation

In this paper, a linearized fully discrete scheme is proposed to solve the two-dimensional nonlinear time fractional Schr\"odinger equation with weakly singular solutions, which is constructed by using L1 scheme for Caputo fractional derivative, backward formula for the approximation of nonlinear term and five-point difference scheme in space. We rigorously prove the unconditional stability and pointwise-in-time convergence of the fully discrete scheme, which does not require any restriction on the grid ratio. Numerical results are presented to verify the accuracy of the theoretical analysis.

math.NA↗

arXiv2024

Hunyuan-Large: An Open-Source MoE Model with 52 Billion Activated Parameters by Tencent

In this paper, we introduce Hunyuan-Large, which is currently the largest open-source Transformer-based mixture of experts model, with a total of 389 billion parameters and 52 billion activation parameters, capable of handling up to 256K tokens. We conduct a thorough evaluation of Hunyuan-Large's superior performance across various benchmarks including language understanding and generation, logical reasoning, mathematical problem-solving, coding, long-context, and aggregated tasks, where it outperforms LLama3.1-70B and exhibits comparable performance when compared to the significantly larger LLama3.1-405B model. Key practice of Hunyuan-Large include large-scale synthetic data that is orders larger than in previous literature, a mixed expert routing strategy, a key-value cache compression technique, and an expert-specific learning rate strategy. Additionally, we also investigate the scaling laws and learning rate schedule of mixture of experts models, providing valuable insights and guidances for future model development and optimization. The code and checkpoints of Hunyuan-Large are released to facilitate future innovations and applications. Codes: https://github.com/Tencent/Hunyuan-Large Models: https://huggingface.co/tencent/Tencent-Hunyuan-Large

cs.CL↗

arXiv2024

MedSAM-U: Uncertainty-Guided Auto Multi-Prompt Adaptation for Reliable MedSAM

The Medical Segment Anything Model (MedSAM) has shown remarkable performance in medical image segmentation, drawing significant attention in the field. However, its sensitivity to varying prompt types and locations poses challenges. This paper addresses these challenges by focusing on the development of reliable prompts that enhance MedSAM's accuracy. We introduce MedSAM-U, an uncertainty-guided framework designed to automatically refine multi-prompt inputs for more reliable and precise medical image segmentation. Specifically, we first train a Multi-Prompt Adapter integrated with MedSAM, creating MPA-MedSAM, to adapt to diverse multi-prompt inputs. We then employ uncertainty-guided multi-prompt to effectively estimate the uncertainties associated with the prompts and their initial segmentation results. In particular, a novel uncertainty-guided prompts adaptation technique is then applied automatically to derive reliable prompts and their corresponding segmentation outcomes. We validate MedSAM-U using datasets from multiple modalities to train a universal image segmentation model. Compared to MedSAM, experimental results on five distinct modal datasets demonstrate that the proposed MedSAM-U achieves an average performance improvement of 1.7\% to 20.5\% across uncertainty-guided prompts.

cs.CV↗

arXiv2024

Textual Inversion and Self-supervised Refinement for Radiology Report Generation

Existing mainstream approaches follow the encoder-decoder paradigm for generating radiology reports. They focus on improving the network structure of encoders and decoders, which leads to two shortcomings: overlooking the modality gap and ignoring report content constraints. In this paper, we proposed Textual Inversion and Self-supervised Refinement (TISR) to address the above two issues. Specifically, textual inversion can project text and image into the same space by representing images as pseudo words to eliminate the cross-modeling gap. Subsequently, self-supervised refinement refines these pseudo words through contrastive loss computation between images and texts, enhancing the fidelity of generated reports to images. Notably, TISR is orthogonal to most existing methods, plug-and-play. We conduct experiments on two widely-used public datasets and achieve significant improvements on various baselines, which demonstrates the effectiveness and generalization of TISR. The code will be available soon.

cs.CV↗

arXiv2024

Diffusion Posterior Proximal Sampling for Image Restoration

Diffusion models have demonstrated remarkable efficacy in generating high-quality samples. Existing diffusion-based image restoration algorithms exploit pre-trained diffusion models to leverage data priors, yet they still preserve elements inherited from the unconditional generation paradigm. These strategies initiate the denoising process with pure white noise and incorporate random noise at each generative step, leading to over-smoothed results. In this paper, we present a refined paradigm for diffusion-based image restoration. Specifically, we opt for a sample consistent with the measurement identity at each generative step, exploiting the sampling selection as an avenue for output stability and enhancement. The number of candidate samples used for selection is adaptively determined based on the signal-to-noise ratio of the timestep. Additionally, we start the restoration process with an initialization combined with the measurement signal, providing supplementary information to better align the generative process. Extensive experimental results and analyses validate that our proposed method significantly enhances image restoration performance while consuming negligible additional computational resources.

eess.IV↗

arXiv2024

X-ray properties of reverberation-mapped AGNs with super-Eddington accreting massive black holes

The X-ray properties of Active Galactic Nuclei (AGNs) depend on their underlying physical parameters, particularly the accretion rate. We identified eight reverberation-mapped AGNs with some of the largest known accretion rates without high-quality X-ray data. We obtained new Chandra ACIS-S X-ray observations and nearly simultaneous optical spectrophotometry to investigate the properties of these AGNs with extreme super-Eddington accreting black holes (SEAMBHs). We combined our new X-ray measurements with those of other reverberation-mapped AGNs, which have the best-determined masses and accretion rates. The trend of the steepening of the spectral slope between X-ray and optical-UV, $\alpha_{\rm ox}$, with increasing optical-UV luminosity, $L_{2500\r{A}}$, holds true for even the most extreme SEAMBHs. One of our new SEAMBHs appears X-ray weak for its luminosity, perhaps due to absorption associated with orientation effects involving a slim disk thought to be present in highly accreting systems. The correlation of the $\rm 2-8~ keV$ X-ray photon index with the accretion rate also holds for the extreme SEAMBHs, which show some of the largest photon indices reported for AGNs.

astro-ph.HE↗

arXiv2023

Proposals for ruling out the real quantum theories in an entanglement-swapping quantum network with causally independent sources

The question of whether complex numbers play a fundamental role in quantum theory has been debated since the inception of quantum mechanics. Recently, a feasible proposal to differentiate between real and complex quantum theories based on the technique of testing Bell nonlocalities has emerged [Nature 600, 625-629 (2021)]. Based on this method, the real quantum theory has been falsified experimentally in both photonic and superconducting quantum systems [Phys. Rev. Lett. 128, 040402 (2022), Phys. Rev. Lett. 128, 040403 (2022)]. The quantum networks with multiple independent sources which are not causally connected have gained significant interest as they offer new perspective on studying the nonlocalities. The independence of these sources imposes additional constraints on observable covariances and leads to new bounds for classical and quantum correlations. In this study, we examine the discrimination between the real and complex quantum theories with an entanglement swapping scenario under a stronger assumption that the two sources are causally independent, which wasn't made in previous works. Using a revised Navascu\'es-Pironio-Ac\'in method and Bayesian optimization, we find a proposal with optimal coefficients of the correlation function which could give a larger discrimination between the real and quantum theories comparing with the existing proposals. This work opens up avenues for further exploration of the discrimination between real and complex quantum theories within intricate quantum networks featuring causally independent parties.

quant-ph↗

arXiv2023

Fast and Stable Diffusion Inverse Solver with History Gradient Update

Diffusion models have recently been recognised as efficient inverse problem solvers due to their ability to produce high-quality reconstruction results without relying on pairwise data training. Existing diffusion-based solvers utilize Gradient Descent strategy to get a optimal sample solution. However, these solvers only calculate the current gradient and have not utilized any history information of sampling process, thus resulting in unstable optimization progresses and suboptimal solutions. To address this issue, we propose to utilize the history information of the diffusion-based inverse solvers. In this paper, we first prove that, in previous work, using the gradient descent method to optimize the data fidelity term is convergent. Building on this, we introduce the incorporation of historical gradients into this optimization process, termed History Gradient Update (HGU). We also provide theoretical evidence that HGU ensures the convergence of the entire algorithm. It's worth noting that HGU is applicable to both pixel-based and latent-based diffusion model solvers. Experimental results demonstrate that, compared to previous sampling algorithms, sampling algorithms with HGU achieves state-of-the-art results in medical image reconstruction, surpassing even supervised learning methods. Additionally, it achieves competitive results on natural images.

cs.CV↗

arXiv2023

Testing Genuine Multipartite Nonlocality via an Inflated Network with Multi-copy Entangled States

Understanding the nonlocality of multipartite quantum systems provides valuable insights into their behaviors and potential applications. In this Letter, assuming a quantum network inflated with multiple copies of genuine multipartite entangled states, we propose a novel noise-robust approach to test the genuine multipartite nonlocality inherent in each copy under Svetlichny's biseparable model. This extends Gisin's Theorem to an arbitrary number of parties, establishing the equivalence among genuine multipartite nonlocality, genuine multipartite steering, and genuine multipartite entanglement for all multipartite pure states under multiple copies assumption. In the experiment, we employ a hybrid photonic quantum network to verify the genuine tripartite nonlocality of generalized Greenberger-Horne-Zeilinger (GHZ) and W states beyond previously explored parameter regimes. This work not only offers a unified robust method on exploring multipartite quantum correlations, but also opens a new avenue for studying genuine multipartite nonlocality through network-distributed multi-copy quantum states and different network topologies.

quant-ph↗