arXiv subjects

Zhixuan Li

Publications and source records attributed to Zhixuan Li.

At least 19 recordsLinked to original sources

arXiv2026

SCOPE-RL: Optimizing Reasoning Paths Before and After Success

Reinforcement learning with verifiable rewards (RLVR) optimizes LLMs using sparse verifiable final-answer rewards. This sparse anchor reliably verifies whether a trajectory succeeds but provides no direct feedback on the reasoning path that produced it. Before success, prerequisite progress on hard problems receives no reward signal; after success, outcome rewards cannot distinguish well-organized correct trajectories from redundant or locally flawed ones. We introduce SCOPE-RL (Scaffolded Chain Optimization with Process Efficiency), a two-stage framework that densifies this anchor while retaining the GRPO update: Adaptive Scaffolded RL adds prefix-decomposed verifiable rewards on answer-hidden sub-question chains before success, and Quality-Aware Process RL applies correctness-gated process-shape rewards to refine correct trajectories after success. An expert-validated Step-Quality Evaluation Protocol evaluates useful-step density, error localization, and token efficiency beyond final-answer accuracy. On Qwen3-8B-Instruct trained on DAPO-Math and Big-Math, SCOPE-RL improves average accuracy by up to 11.2 pp and reduces reasoning tokens by up to 27.1% over outcome-only GRPO; the gains hold under GSPO and on Qwen3-0.6B-Instruct, indicating that reward-signal densification is complementary to policy-update-level RLVR advances. Code and data are available at https://github.com/tokencraft-lab/SCOPE-RL.

cs.LG↗

arXiv2026

STAMP: Provenance-Guided Credit Assignment for Deep Search Agents

Reinforcement learning for deep-search agents has largely focused on trajectory-level scoring -- outcome correctness, citation-aware rewards, and evidence coverage. Yet the actions that expose supporting documents receive no targeted credit, a gap we call the reward-credit mismatch. We propose STAMP, in which a reference-based verifier judges whether each cited document supports an entity or relation in a training-time evidence graph, and first-exposure attribution traces each supported citation back to the action that first surfaced it. This step credit is injected through sign-preserving advantage modulation, which redistributes advantage across steps without changing the trajectory-level reward or the relative ranking of trajectories within each group. On BrowseComp, BrowseComp-ZH, and xbench-DS, STAMP improves the GRPO baseline by +2.0/+5.5/+3.0 points under matched SFT initialization, training data, and search tools, and composes with both outcome-only and citation-rubric base rewards. Component ablations confirm that the provenance-based credit signal and the sign-preserving advantage modulation each contribute to the gains.

cs.AI↗

arXiv2026

Enhancing VLBI Capability with the SKA-Mid and the Jingdong 120-m Radio Telescope

The Jingdong Radio Telescope (JRT) is a 120-meter fully steerable radio telescope currently under construction in Jingdong County, Yunnan Province, China. Located at a relatively low latitude (24.5 degree), the JRT will enable observations of nearly 90% of the sky. Equipped with two broadband single-pixel receivers covering 1-8 GHz and 6-18 GHz, and a powerful digital backend, the telescope will support single-dish studies of various radio sources-particularly millisecond pulsars for enhancing the detection of nanohertz gravitational waves. In addition to single-dish capabilities, the JRT is expected to contribute approximately 800 hours annually to international Very Long Baseline Interferometry (VLBI) observations via a standard VLBI backend. When operating in conjunction with the phased-up SKA-Mid, the JRT will significantly enhance the technical and scientific capabilities of existing VLBI networks. This paper presents a comprehensive overview of the JRT's VLBI module and explores its potential to improve joint VLBI observations with current VLBI networks. Our analysis suggests that coordinated VLBI observations involving both the SKA-Mid and the JRT have the potential to significantly advance the field. For early sciences, we also highlight a few highly promising scientific cases, e.g. measuring the distance to PSR J0437-4715 with <1 ly accuracy and exploring jet formation with an event-horizon-scale resolution in M60*.

astro-ph.IM↗

arXiv2026

Data and Evaluation Closed-Loop for Model Capability Enhancement

Model capability is the central variable in LLM pre-training, yet is never observed directly: data shapes it prospectively, while evaluation reveals it only retrospectively, compressing samples, prompts, decoding, and scoring rules into one noisy score. Practical optimization runs this backward: a failure is observed first, and the engineer must infer the corpus fix. The two sides speak incompatible vocabularies -- benchmark names and per-sample correctness versus data sources, domains, and quality labels -- so this inference is usually intuition, not method. We close this gap with the \emph{capability slice}: a group of evaluation samples sharing background condition, task type, solving operation, and output constraint -- precise enough to localize a single weakness yet stable enough to survive aggregation, unlike a benchmark name, too coarse, or a single sample, too noisy. Built around this unit, an evaluation taxonomy, a non-instruction data taxonomy, and mapping rules form a closed loop turning a benchmark-level failure into a targeted, testable data intervention. We test this loop on two case studies pulling in opposite directions. First, the loop rules the data out: continued pre-training drives BBH down by $-46.82\%$, but diagnosis traces this to a single masked \texttt{\textless EOS\textgreater} loss rather than weakened reasoning; restoring it recovers BBH to $66.44$, above the original checkpoint, without changing the data. Second, the loop rules the data in: a persistent math-reasoning weakness is decomposed by solving operation into specific failing combinations, and a weakness-targeted sampling procedure built from it lifts AIME2025/AIME2026 Pass@128 from $6.67$/$0.00$ to $26.67$ each. The same unmodified loop reaches opposite, correct verdicts in both cases, showing the evaluation-to-data inference can be routine, auditable, and experimentally validated rather than intuitive.

cs.AI↗

arXiv2026

Let the Data Decide: Supervision Analysis, Capability Trade-offs, and Adaptive Objective Routing in Continued Pre-Training via Off-Policy Distillation

Off-policy distillation is now central to large language model pre-training, yet how training data, objective parameterization, and model capabilities interact remains poorly characterized. We studies top-$k$-truncated, temperature-scaled off-policy distillation by decomposing this problem into two questions: an \emph{objective-to-capability} analysis of how the training objective shapes token-level supervision and downstream performance, and a \emph{data-to-objective} analysis of how data heterogeneity should inform objective routing. We first show that the language-modeling objective ($L_{\mathrm{LM}}$) and the knowledge-distillation objective ($L_{\mathrm{KD}}$) induce systematically different capability profiles, and trace this divergence to a gradient-level tension between \emph{direct observed-token reinforcement} and \emph{teacher-supported alternative supervision}. To quantify this tension, we introduce diagnostic metrics -- support coverage, observed-token probability mass, and teacher-distribution concentration -- and show via controlled sweeps that the support size $k$ governs a coverage-sharpness trade-off, while distillation temperature controls within-support probability allocation. We then examine adaptive objective routing: a domain-level policy that applies $L_{\mathrm{LM}}$ to math and code and $L_{\mathrm{KD}}$ to general-domain data yields consistent gains over both single-objective baselines, whereas token-level routing based on observed-token probability mass or teacher entropy fails to consistently match the single-objective baseline. These results suggest that effective objective routing depends less on routing granularity than on the quality of the routing signal, reframing continued pre-training via off-policy distillation as a structured, data-conditional supervision-design problem rather than a global hyperparameter choice.

cs.LG↗

arXiv2026

Discovery of a nonsymmorphic superconductor with spontaneous rotational symmetry breaking and nontrivial zero modes

Topological superconductivity has attracted great interest due to its fundamental significance for realizing Majorana quasiparticles and fault-tolerant quantum computation. Nonsymmorphic superconductors, with symmetry-protected nontrivial electronic structures, offer a promising route to exotic topological superconducting states, yet experimental realizations remain scarce. Here we identify nonsymmorphic compound PtPb4 as a robust platform hosting superconductivity with spontaneous rotational symmetry breaking and nontrivial zero-energy modes. PtPb4 crystallizes in a frustrated Shastry-Sutherland lattice and exhibits nontrivial band topology. By combining in-plane and out-of-plane resistivity measurements, pronounced twofold anisotropy is observed in both the superconducting state and the upper critical field, evidencing spontaneous rotational symmetry breaking. Scanning tunneling microscopy/spectroscopy further reveal twofold-symmetric magnetic vortices, providing direct real-space evidence for the symmetry-broken superconducting state. Notably, a robust zero-energy vortex bound state emerges and persists without spatial splitting over extended distances, consistent with the characteristics expected for Majorana bound state. These findings uncover an exotic superconducting state in PtPb4 and establish a promising platform for exploring topological superconductivity and superconducting quantum devices.

cond-mat.supr-con↗

arXiv2026

DipGuava: Disentangling Personalized Gaussian Features for 3D Head Avatars from Monocular Video

While recent 3D head avatar creation methods attempt to animate facial dynamics, they often fail to capture personalized details, limiting realism and expressiveness. To fill this gap, we present DipGuava (Disentangled and Personalized Gaussian UV Avatar), a novel 3D Gaussian head avatar creation method that successfully generates avatars with personalized attributes from monocular video. DipGuava is the first method to explicitly disentangle facial appearance into two complementary components, trained in a structured two-stage pipeline that significantly reduces learning ambiguity and enhances reconstruction fidelity. In the first stage, we learn a stable geometry-driven base appearance that captures global facial structure and coarse expression-dependent variations. In the second stage, the personalized residual details not captured in the first stage are predicted, including high-frequency components and nonlinearly varying features such as wrinkles and subtle skin deformations. These components are fused via dynamic appearance fusion that integrates residual details after deformation, ensuring spatial and semantic alignment. This disentangled design enables DipGuava to generate photorealistic, identity-preserving avatars, consistently outperforming prior methods in both visual quality and quantitativeperformance, as demonstrated in extensive experiments.

cs.CV↗

arXiv2026

CosmicWeb-21cm array: A New Radio Observation Array Design for 21cm Cosmology

This paper presents the CosmicWeb-21cm array, a novel radio interferometer designed to overcome the key challenges in 21 cm cosmology. Its core innovations include: (1) a multi-scale nested geometry combining a hexagonal core with logarithmic spiral arms for excellent UV coverage and calibration robustness; (2) an intelligent non-uniform frequency sampling strategy that adapts resolution to foreground and signal characteristics, reducing data volume while preserving information; and (3) a machine-learning-enhanced, physics-informed processing pipeline that achieves 99.7\% foreground removal efficiency; (4) a dual-polarization crossed dipole integrated with a dielectric lens and cryogenically cooled LNA, achieving stable beam patterns and low noise temperature ($<35$ K) across 50-250 MHz. These co-designed advances enable high sensitivity mapping of the Epoch of Reionization, dark energy constraints and cosmic-web structure.

astro-ph.IM↗

arXiv2026

SEGA: A Transferable Signed Ensemble Gaussian Black-Box Attack against No-Reference Image Quality Assessment Models

No-Reference Image Quality Assessment (NR-IQA) models play an important role in various real-world applications. Recently, adversarial attacks against NR-IQA models have attracted increasing attention, as they provide valuable insights for revealing model vulnerabilities and guiding robust system design. Some effective attacks have been proposed against NR-IQA models in white-box settings, where the attacker has full access to the target model. However, these attacks often suffer from poor transferability to unknown target models in more realistic black-box scenarios, where the target model is inaccessible. This work makes the first attempt to address the challenge of low transferability in attacking NR-IQA models by proposing a transferable Signed Ensemble Gaussian black-box Attack (SEGA). The main idea is to approximate the gradient of the target model by applying Gaussian smoothing to source models and ensembling their smoothed gradients. To ensure the imperceptibility of adversarial perturbations, SEGA further removes inappropriate perturbations using a specially designed perturbation filter mask. Experimental results on the CLIVE dataset demonstrate the superior transferability of SEGA, validating its effectiveness in enabling successful transfer-based black-box attacks against NR-IQA models.

cs.CV↗

arXiv2026

A Fourth-Order Cut-cell Multigrid Method for Solving Elliptic Equations on Arbitrary Domains

To numerically solve a generic elliptic equation on two-dimensional domains with rectangular Cartesian grids, we propose a cut-cell geometric multigrid method that features (1) general algorithmic steps that apply to two-dimensional constant-coefficient elliptic equations with both divergence and non-divergence forms and all types of boundary conditions, (2) the versatility of handling both regular and irregular domains with arbitrarily complex topology and geometry, (3) the fourth-order accuracy even at the presence of ${\cal C}^1$ discontinuities on the domain boundary, and (4) the optimal complexity of $O(h^{-2})$.Test results demonstrate the generality, accuracy, efficiency, robustness, and excellent conditioning of the proposed method.

math.NA↗

arXiv2025

Q-CLIP: Unleashing the Power of Vision-Language Models for Video Quality Assessment through Unified Cross-Modal Adaptation

Accurate and efficient Video Quality Assessment (VQA) has long been a key research challenge. Current mainstream VQA methods typically improve performance by pretraining on large-scale classification datasets (e.g., ImageNet, Kinetics-400), followed by fine-tuning on VQA datasets. However, this strategy presents two significant challenges: (1) merely transferring semantic knowledge learned from pretraining is insufficient for VQA, as video quality depends on multiple factors (e.g., semantics, distortion, motion, aesthetics); (2) pretraining on large-scale datasets demands enormous computational resources, often dozens or even hundreds of times greater than training directly on VQA datasets. Recently, Vision-Language Models (VLMs) have shown remarkable generalization capabilities across a wide range of visual tasks, and have begun to demonstrate promising potential in quality assessment. In this work, we propose Q-CLIP, the first fully VLMs-based framework for VQA. Q-CLIP enhances both visual and textual representations through a Shared Cross-Modal Adapter (SCMA), which contains only a minimal number of trainable parameters and is the only component that requires training. This design significantly reduces computational cost. In addition, we introduce a set of five learnable quality-level prompts to guide the VLMs in perceiving subtle quality variations, thereby further enhancing the model's sensitivity to video quality. Furthermore, we investigate the impact of different frame sampling strategies on VQA performance, and find that frame-difference-based sampling leads to better generalization performance across datasets. Extensive experiments demonstrate that Q-CLIP exhibits excellent performance on several VQA datasets.

cs.CV↗

arXiv2025

UGD-IML: A Unified Generative Diffusion-based Framework for Constrained and Unconstrained Image Manipulation Localization

In the digital age, advanced image editing tools pose a serious threat to the integrity of visual content, making image forgery detection and localization a key research focus. Most existing Image Manipulation Localization (IML) methods rely on discriminative learning and require large, high-quality annotated datasets. However, current datasets lack sufficient scale and diversity, limiting model performance in real-world scenarios. To overcome this, recent studies have explored Constrained IML (CIML), which generates pixel-level annotations through algorithmic supervision. However, existing CIML approaches often depend on complex multi-stage pipelines, making the annotation process inefficient. In this work, we propose a novel generative framework based on diffusion models, named UGD-IML, which for the first time unifies both IML and CIML tasks within a single framework. By learning the underlying data distribution, generative diffusion models inherently reduce the reliance on large-scale labeled datasets, allowing our approach to perform effectively even under limited data conditions. In addition, by leveraging a class embedding mechanism and a parameter-sharing design, our model seamlessly switches between IML and CIML modes without extra components or training overhead. Furthermore, the end-to-end design enables our model to avoid cumbersome steps in the data annotation process. Extensive experimental results on multiple datasets demonstrate that UGD-IML outperforms the SOTA methods by an average of 9.66 and 4.36 in terms of F1 metrics for IML and CIML tasks, respectively. Moreover, the proposed method also excels in uncertainty estimation, visualization and robustness.

cs.CV↗

arXiv2025

BEAT: Balanced Frequency Adaptive Tuning for Long-Term Time-Series Forecasting

Time-series forecasting is crucial for numerous real-world applications including weather prediction and financial market modeling. While temporal-domain methods remain prevalent, frequency-domain approaches can effectively capture multi-scale periodic patterns, reduce sequence dependencies, and naturally denoise signals. However, existing approaches typically train model components for all frequencies under a unified training objective, often leading to mismatched learning speeds: high-frequency components converge faster and risk overfitting, while low-frequency components underfit due to insufficient training time. To deal with this challenge, we propose BEAT (Balanced frEquency Adaptive Tuning), a novel framework that dynamically monitors the training status for each frequency and adaptively adjusts their gradient updates. By recognizing convergence, overfitting, or underfitting for each frequency, BEAT dynamically reallocates learning priorities, moderating gradients for rapid learners and increasing those for slower ones, alleviating the tension between competing objectives across frequencies and synchronizing the overall learning process. Extensive experiments on seven real-world datasets demonstrate that BEAT consistently outperforms state-of-the-art approaches.

cs.LG↗

arXiv2025

Single Point, Full Mask: Velocity-Guided Level Set Evolution for End-to-End Amodal Segmentation

Amodal segmentation aims to recover complete object shapes, including occluded regions with no visual appearance, whereas conventional segmentation focuses solely on visible areas. Existing methods typically rely on strong prompts, such as visible masks or bounding boxes, which are costly or impractical to obtain in real-world settings. While recent approaches such as the Segment Anything Model (SAM) support point-based prompts for guidance, they often perform direct mask regression without explicitly modeling shape evolution, limiting generalization in complex occlusion scenarios. Moreover, most existing methods suffer from a black-box nature, lacking geometric interpretability and offering limited insight into how occluded shapes are inferred. To deal with these limitations, we propose VELA, an end-to-end VElocity-driven Level-set Amodal segmentation method that performs explicit contour evolution from point-based prompts. VELA first constructs an initial level set function from image features and the point input, which then progressively evolves into the final amodal mask under the guidance of a shape-specific motion field predicted by a fully differentiable network. This network learns to generate evolution dynamics at each step, enabling geometrically grounded and topologically flexible contour modeling. Extensive experiments on COCOA-cls, D2SA, and KINS benchmarks demonstrate that VELA outperforms existing strongly prompted methods while requiring only a single-point prompt, validating the effectiveness of interpretable geometric modeling under weak guidance. The code will be publicly released.

cs.CV↗

arXiv2025

Shape Distribution Matters: Shape-specific Mixture-of-Experts for Amodal Segmentation under Diverse Occlusions

Amodal segmentation targets to predict complete object masks, covering both visible and occluded regions. This task poses significant challenges due to complex occlusions and extreme shape variation, from rigid furniture to highly deformable clothing. Existing one-size-fits-all approaches rely on a single model to handle all shape types, struggling to capture and reason about diverse amodal shapes due to limited representation capacity. A natural solution is to adopt a Mixture-of-Experts (MoE) framework, assigning experts to different shape patterns. However, naively applying MoE without considering the object's underlying shape distribution can lead to mismatched expert routing and insufficient expert specialization, resulting in redundant or underutilized experts. To deal with these issues, we introduce ShapeMoE, a shape-specific sparse Mixture-of-Experts framework for amodal segmentation. The key idea is to learn a latent shape distribution space and dynamically route each object to a lightweight expert tailored to its shape characteristics. Specifically, ShapeMoE encodes each object into a compact Gaussian embedding that captures key shape characteristics. A Shape-Aware Sparse Router then maps the object to the most suitable expert, enabling precise and efficient shape-aware expert routing. Each expert is designed as lightweight and specialized in predicting occluded regions for specific shape patterns. ShapeMoE offers well interpretability via clear shape-to-expert correspondence, while maintaining high capacity and efficiency. Experiments on COCOA-cls, KINS, and D2SA show that ShapeMoE consistently outperforms state-of-the-art methods, especially in occluded region segmentation. The code will be released.

cs.CV↗

arXiv2025

Unveiling the Invisible: Reasoning Complex Occlusions Amodally with AURA

Amodal segmentation aims to infer the complete shape of occluded objects, even when the occluded region's appearance is unavailable. However, current amodal segmentation methods lack the capability to interact with users through text input and struggle to understand or reason about implicit and complex purposes. While methods like LISA integrate multi-modal large language models (LLMs) with segmentation for reasoning tasks, they are limited to predicting only visible object regions and face challenges in handling complex occlusion scenarios. To address these limitations, we propose a novel task named amodal reasoning segmentation, aiming to predict the complete amodal shape of occluded objects while providing answers with elaborations based on user text input. We develop a generalizable dataset generation pipeline and introduce a new dataset focusing on daily life scenarios, encompassing diverse real-world occlusions. Furthermore, we present AURA (Amodal Understanding and Reasoning Assistant), a novel model with advanced global and spatial-level designs specifically tailored to handle complex occlusions. Extensive experiments validate AURA's effectiveness on the proposed dataset.

cs.CV↗

arXiv2025

LVPNet: A Latent-variable-based Prediction-driven End-to-end Framework for Lossless Compression of Medical Images

Autoregressive Initial Bits is a framework that integrates sub-image autoregression and latent variable modeling, demonstrating its advantages in lossless medical image compression. However, in existing methods, the image segmentation process leads to an even distribution of latent variable information across each sub-image, which in turn causes posterior collapse and inefficient utilization of latent variables. To deal with these issues, we propose a prediction-based end-to-end lossless medical image compression method named LVPNet, leveraging global latent variables to predict pixel values and encoding predicted probabilities for lossless compression. Specifically, we introduce the Global Multi-scale Sensing Module (GMSM), which extracts compact and informative latent representations from the entire image, effectively capturing spatial dependencies within the latent space. Furthermore, to mitigate the information loss introduced during quantization, we propose the Quantization Compensation Module (QCM), which learns the distribution of quantization errors and refines the quantized features to compensate for quantization loss. Extensive experiments on challenging benchmarks demonstrate that our method achieves superior compression efficiency compared to state-of-the-art lossless image compression approaches, while maintaining competitive inference speed. The code is at https://github.com/scy-Jackel/LVPNet.

eess.IV↗

arXiv2025

MS-IQA: A Multi-Scale Feature Fusion Network for PET/CT Image Quality Assessment

Positron Emission Tomography / Computed Tomography (PET/CT) plays a critical role in medical imaging, combining functional and anatomical information to aid in accurate diagnosis. However, image quality degradation due to noise, compression and other factors could potentially lead to diagnostic uncertainty and increase the risk of misdiagnosis. When evaluating the quality of a PET/CT image, both low-level features like distortions and high-level features like organ anatomical structures affect the diagnostic value of the image. However, existing medical image quality assessment (IQA) methods are unable to account for both feature types simultaneously. In this work, we propose MS-IQA, a novel multi-scale feature fusion network for PET/CT IQA, which utilizes multi-scale features from various intermediate layers of ResNet and Swin Transformer, enhancing its ability of perceiving both local and global information. In addition, a multi-scale feature fusion module is also introduced to effectively combine high-level and low-level information through a dynamically weighted channel attention mechanism. Finally, to fill the blank of PET/CT IQA dataset, we construct PET-CT-IQA-DS, a dataset containing 2,700 varying-quality PET/CT images with quality scores assigned by radiologists. Experiments on our dataset and the publicly available LDCTIQAC2023 dataset demonstrate that our proposed model has achieved superior performance against existing state-of-the-art methods in various IQA metrics. This work provides an accurate and efficient IQA method for PET/CT. Our code and dataset are available at https://github.com/MS-IQA/MS-IQA/.

eess.IV↗