arXiv subjects

Xin Wang

Publications and source records attributed to Xin Wang.

At least 127 records · Page 7Linked to original sources

arXiv2026

OmniClimate-TC: Physics-Aware Visual Abstractions for Multimedia Reasoning over Tropical Cyclones

Meteorological reanalysis encodes extreme weather through continuous, physically constrained fields, posing a fundamental challenge for vision-language models (VLMs) whose perceptual assumptions are shaped by natural images. Tropical cyclones exemplify this mismatch: critical properties such as intensity extrema, asymmetry, spatial extent, and physical impacts arise from field-level organization rather than object-centric visual cues. Existing approaches address this gap through text alignment or annotation, treating the problem as multimodal supervision rather than representation design. We introduce Physics-Aware Visual Abstraction (PAVA), a plug-and-play physics-aware representation and annotation interface that maps physical reanalysis fields to visually identifiable and semantically grounded perceptual abstractions for supervision and evaluation in vision-language reasoning. Building on PAVA, we construct OmniClimate-TC, a benchmark for tropical cyclone analysis spanning five classes of reasoning and nine tasks, with 243,890 physically grounded instruction-tuning pairs. Using PAVA-aligned supervision, we adapt VLMs and provide evidence that this representation design improves reasoning over tropical cyclone hazard fields. Our results position OmniClimate-TC as a benchmark for multimedia reasoning over structured geophysical fields, and highlight representation design as a key ingredient for physically grounded reasoning in scientific media.

cs.CE↗

arXiv2026

From localized dryout to convective elongated vapor structures: Reynolds number effects on boiling transition in a rectangular mini-channel

Three-dimensional conjugate simulations were conducted to investigate saturated flow boiling in a rectangular mini-channel, with particular emphasis on the role of inlet Reynolds number on boiling mode selection and transition. A C++ based open-source numerical framework was employed, incorporating a physically informed multi-site nucleation model by coupling a nucleation site density correlation with a Halton-sequence based spatial allocation strategy. Two distinct Re-dependent transition pathways were identified. At low Re, boiling transition is mainly associated with localized dryout development associated with upstream active boiling and progressive downstream liquid starvation. At high Re, the transition is characterized by convective stretching and reorganization of vapor structures, through which elongated vapor slugs evolve into localized vapor films and eventually approach full surface vapor coverage. The global heat transfer characteristics and peak heat transfer capacity are further interpreted in conjunction with boiling mode transition, clarifying the respective roles of wall dryout and volumetric vapor fraction in heat transfer deterioration. Among all cases, Re=2000 provides the most favorable overall thermal response. Overall, within the rectangular mini-channel configuration and operating range considered in this study, Re is closely associated with vapor organization, boiling transition, wall dryout, and global heat transfer performance.

physics.flu-dyn↗

arXiv2026

Artificial Intelligence in a Photonic Temporal Processor

Optical neural networks (ONNs) promise high-throughput and energy-efficient artificial intelligence, yet essentially all implementations so far encode information across space either in free-space arrays or in integrated waveguide meshes, tying the number of neurons to the number of physical components and fixes the routing topology at fabrication. Here we show that moving the computation into time decouples computational dimension from hardware dimension. Exploiting space-time duality, we implement optical diffraction and interference entirely in time domain, using thin-film lithium niobate modulators as time lenses and temporal masks, with chromatic dispersion providing the coupling between successive temporal neurons. We experimentally verify high-order, complex-valued matrix-matrix multiplications using just a single optical input/output port, scaling the computational dimensions far beyond the channel count. By incorporating optical feedback, we extend this platform into versatile neural networks, where the network layers, neuron numbers, and synaptic connections are fully programmable and in-situ trainable. Our temporal diffractive neural networks are successfully validated on various classification benchmarks, alongside image and video generation tasks. Notably, using this platform we demonstrate an all-analogue generative pipeline in which the latent variable is drawn directly from amplified spontaneous emission, so that no digital sampling or electronic modulation appears anywhere in the generative path. Furthermore, high-resolution images and videos are generated at high frame rates, outperforming state-of-the-art modulator-refresh-limited optical generative systems. These results establish a unified photonic temporal computing framework, providing a scalable and deployable pathway toward next-generation machine intelligence.

physics.optics↗

arXiv2026

Deciphering Matter Invariants via Renormalization Group Equations for Neutrino Oscillations

We utilize renormalization group equations (RGEs) for neutrino oscillations in matter to decipher the structure of exact matter invariants. By combining the RGEs with the $S^{}_3$ permutation covariance under relabeling of the neutrino mass eigenstates, we recast all five algebraically independent matter invariants in the three-flavor framework as exact first integrals. Treating the matter potential as a matter spurion and imposing the cancellation conditions for the $1/\widetildeΔ_{ij}^{}$ poles in the RGEs, we further prove that the three-flavor framework contains only two independent monomial invariants, which can be related to the Naumov and Toshev relations. We then extend the analysis to the four-flavor framework, where we uncover a complete set of eleven algebraically independent matter invariants and prove that the rank-two electron--sterile spurion obstructs the common pole cancellation required for any nontrivial multiplicative monomial invariant.

hep-ph↗

arXiv2026

HOLMES: In-Context Failure-Center Localization for High-Dimensional Yield Estimation

Importance sampling for high-sigma yield estimation requires locating the failure center from a severely imbalanced sample set. Existing surrogate-assisted methods rely on iterative gradient-based training, ill-posed under extreme class imbalance; model errors propagate into the estimator, causing accuracy collapse in high dimensions. We recast failure-center localization as few-shot binary classification: a prior-fitted tabular foundation model performs gradient-free in-context inference in a single forward pass, eliminating the ill-posed training loop. \textbf{HOLMES} (High-sigma Optimal Localization via Manifold Estimation and Sampling) pairs this with an SVD-based anisotropic proposal that captures the local geometry of the failure manifold, and a hit-rate-driven adaptive mixing scheme that stabilizes importance weights where conventional adaptation collapses. On 6T SRAM benchmarks spanning $D = 108$ to $D = 1{,}152$, full-dimensional baselines exhibit accuracy collapse at some dimension, with the strongest baseline reaching 25.8\% relative error; PCA+MNIS is additionally evaluated at the two largest dimensions. HOLMES remains within 5.9\% across all five configurations with up to $58.8\times$ speedup over Monte Carlo. The code is available on \href{https://github.com/IceLab-JCIE/ICE006-Yield-Holmes}

cs.AR↗

arXiv2026

Spectral-Efficient MIMO-OFDM: Low-Complexity Solution based on Random Multiplexing

This paper presents a low-complexity precoded MIMO-OFDM system for achieving improved spectral efficiency (SE) via intentionally compressing information symbols among subcarriers. Particularly, the proposed scheme leverages the powerful random multiplexing mechanism for precoding, and adopts the linear-complexity orthogonal approximate message passing (OAMP) estimator for symbol detection, where the compatibility with the existing fifth generation (5G) architectures is fully preserved. We further provide the theoretical analysis based on the replica-symmetric (RS) formula. This analysis confirms the advantages of the proposed system with respect to the adopted compression ratios, where an interesting phase transition behavior is verified. Numerical results coincide with our analysis and demonstrate significant improvements in terms of achievable rates and bit error rate (BER) compared to conventional MIMO-OFDM counterpart, making the proposed scheme a promising solution to 6G and beyond wireless networks.

eess.SP↗

arXiv2026

Graph Engineering in the Era of LLM Agents: From Individual Intelligence to System Intelligence

LLMs have evolved from language generators to autonomous agents capable of complex, long-horizon tasks. This evolution has produced paradigms including Prompt Engineering to elicit model capabilities, Context Engineering to manage information access, Harness Engineering to organize external tools and resources, and Loop Engineering to support continual reflection and self-improvement. Yet as tasks grow more complex, individual intelligence faces a fundamental limit: many tasks require heterogeneous expertise, interdependent subtasks, parallel execution, independent verification, and persistent state, exceeding any single agent's organizational capacity. Augmenting one agent's capabilities or context cannot resolve this architectural mismatch; intelligence must instead be distributed across specialized agents and organized at the system level. We call this System Intelligence: an agent system's ability to organize and coordinate multiple intelligent components into a coherent, adaptive whole pursuing a shared objective. Achieving it requires more than adding agents; it demands explicit structures to organize work, coordinate heterogeneous agents, and maintain evolving execution states. We introduce Graph Engineering, an emerging paradigm for next-generation agent systems. Unlike prior paradigms that mainly optimize individual interactions or agent-level behavior, Graph Engineering constructs explicit, dynamic, evolving graph structures representing tasks, agents, and system states. These abstractions provide a unified foundation for organizing complex objectives, orchestrating heterogeneous agents, modeling system dynamics, and enabling scalable agent evolution. We systematically review the principles, methodologies, and applications of Graph Engineering for LLM agents. Related papers, open-source data, and projects are collected at https://github.com/DEEP-JLU/Awesome-Graph-Engineering.

cs.IR↗

arXiv2026

Combining Self-Embedding Audio Watermarking with Ultra-Low-Bitrate Neural Codecs

Partial manipulation of speech recordings, where only localized segments of an utterance are altered, poses a significant challenge for content integrity verification, as reliable detection and localization of such edits becomes harder as the manipulated proportion decreases. Watermarking offers a proactive defense alternative by embedding auxiliary information prior to distribution; classical hash-based schemes achieve near-perfect detection and localization under ideal conditions, but the original content cannot be recovered once a segment is manipulated. Building on a prior self-embedding audio steganography framework, this work presents an initial exploration of proactive defense performance under ideal conditions, extending the investigation along three axes: frame-level localization, multi-bit least significant bit variants, and evaluation across multiple ultra-low-bitrate neural codec representations. By embedding a compact neural codec representation rather than a cryptographic hash, the framework additionally enables recovery of the manipulated regions, while supporting training-free detection and localization without spoofed examples. Experiments across four controlled manipulation types under ideal channel conditions show that the embedded payload, and hence an approximate reconstruction of the authentic content, is always fully recovered without bit errors. The results also indicate that the choice of neural codec is the dominant factor for detection and localization performance.

cs.SD↗

arXiv2026

ReLog: Execution-Aware Logging with Runtime Feedback for LLM-Oriented Debugging

Logging statements are essential for software debugging and maintenance. However, existing approaches to automatic logging generation rely on static analysis and produce statements in a single pass without considering runtime behavior. They are also typically evaluated by similarity to developer-written logs, assuming these logs form an adequate gold standard. This assumption is increasingly limiting in the LLM era, where logs are consumed not only by developers but also by LLMs for downstream tasks. As a result, optimizing logs for human similarity does not necessarily reflect their practical utility. To address these limitations, we introduce ReLog, an iterative logging generation framework guided by runtime feedback. ReLog leverages LLMs to generate, execute, evaluate, and refine logging statements so that runtime logs better support downstream tasks. Instead of comparing against developer-written logs, we evaluate ReLog through downstream debugging tasks, including defect localization and repair. We construct a benchmark based on Defects4J under both direct and indirect debugging settings. Results show that ReLog consistently outperforms all baselines, achieving an F1 score of 0.520 and repairing 97 defects in the direct setting, and the best F1 score of 0.408 in the indirect setting where source code is unavailable. Additional experiments across multiple LLMs demonstrate the generality of the framework, while ablations confirm the importance of iterative refinement and compilation repair. Overall, our work reframes logging as a runtime-guided, task-oriented process and advocates evaluating logs by their downstream utility rather than textual similarity.

cs.SE↗

arXiv2026

AgentSpec: Speculative Decoding for Batch Inference of LLM Agents

Large language model (LLM)-based agent applications often incur high response time. Speculative decoding is a promising solution to improve the inference efficiency of LLM agents without impacting generation quality. However, state-of-the-art speculative decoding algorithms exhibit substantial speed degradation under large batch sizes, limiting their effectiveness to deploy in real-world agent applications. In this work, we first present a systematic analysis of speculative decoding for LLM agents and identify two dominant factors of speedup degradation: high rejection rate of speculative tokens, and under-utilization of dynamic token budgets.B ased on these observations, we propose AgentSpec, a speculative decoding algorithm that addresses the limitations of existing methods for LLM agents. AgentSpec incorporates structure-isolated drafting that constrains speculation to semantically coherent segments of the agent workflow, reducing the drafts of irrelevant semantic paths and achieving an extremely low rejection rate. Moreover, AgentSpec adopts redundancy-aware budget allocation that exploits agent-level information to better utilize the dynamically-free token budget during the agent inference. We implement and evaluate AgentSpec on five different workloads and four different models from four different LLM families in vLLM. Our results demonstrate the superiority of AgentSpec over state-of-the-arts.

cs.CL↗

arXiv2026

Improving Cross-Site Whole-Heart Segmentation

Whole-heart segmentation from CT and MRI is essential for quantitative cardiac image analysis, but remains challenging under multi-center and multi-modality distribution shift. In the CARE whole-heart segmentation task, models must generalize from limited labeled sites to unseen acquisition distributions, where variation in spacing, intensity, reconstruction texture, and anatomy can degrade out-of-distribution performance. We propose a modality-routed 3D cardiac segmentation pipeline that combines TotalSegmentator-initialized nnU-Netv2 models with site-characterized, label-preserving appearance augmentation. We first characterize the available sites using measurable image properties and use this analysis to motivate candidate data-space generalization routes. The final retained recipe applies Bias Field + Bezier appearance augmentation, combining smooth spatial intensity perturbation with nonlinear intensity remapping, followed by lightweight class-wise largest-connected-component cleanup. On the primary held-out-site validation splits, the final configuration improves CT mean Dice from 0.8350 to 0.9135 and MRI mean Dice from 0.7695 to 0.7830, while also reducing HD95. These results suggest that site-motivated appearance augmentation is a practical strategy for improving cross-site robustness in limited-data whole-heart segmentation. Our code can be found in https://github.com/Purdue-M2/Improving-Cross-Site-Whole-Heart-Segmentation

eess.IV↗

arXiv2026

Chiral doublon bound states in a synthetic topological waveguide

Nonlinear waveguide QED serves as a platform for studying correlated photon dynamics, where its interplay with topological bound states can give rise to unconventional phenomena. We investigate a triangular ladder waveguide in which a pair of emitters forms a doublon bound state through strong nearest-neighbor interactions. By mapping the doublon dynamics onto an effective SSH chain, we show that the system is equivalent to an effective emitter coupled to a topological doublon bath. Using the vacancy-like dressed state approach, we demonstrate the existence of a chiral bound state with a node at the coupling site and an exponentially localized wave function inherited from the SSH edge states. Extending this picture to two emitter pairs, we derive an effective four-body interaction whose strength is determined by the relative chirality and the separation parity of the bound states. When the two bound states face each other, coherent Rabi oscillations emerge; when they are arranged back-to-back or have an even separation, the interaction vanishes. Our results provide a route toward engineering tunable many-body interactions between correlated photon pairs and realizing nonlinear quantum networks based on flying doublons.

quant-ph↗

arXiv2026

Loopy: Seamless Video Loop Generation via Anchored Looping Shift of Positional Embedding

Looping videos are essential for practical applications such as web graphics, game development, and social media. However, existing approaches typically fail to generate high-quality looping videos due to the neglect of how video generation models perceive temporal order and how this relates to the looping behavior. In this work, we are the first to reveal that position embedding at different attention layers within DiT exhibits varying levels of positional control, with the most pronounced layer acting as an anchor. We formulate this anchored layer as the reference point of the looping video, offering strong contextual priors for the remaining layers to facilitate the generation of seamless and coherent video content. Based on this insight, we propose an anchored position embedding shifting strategy that applies layer-specific shift lengths according to each layer's temporal control effect, effectively transforming DiT's temporal perception from a straight line to a circle. Leveraging this strategy, we develop a general framework, Loopy, for high-quality looping video generation, supporting both RGB and RGBA videos, while also enabling advanced AIGC features such as identity control and style transfer. Experiments demonstrate that our approach significantly improves temporal consistency and visual fidelity in generated looping videos. The released model is available on our website: https://donghaotian123.github.io/Loopy.

cs.CV↗

arXiv2026

Learning Spatially Adaptive Structural Coordination for Underwater Salient Object Detection

Underwater salient object detection (USOD) has attracted increasing attention for underwater scene understanding and vision-guided robotic applications. However, the spatially non-uniform degradation in underwater images causes spatially varying reliability of structural cues: boundary-sensitive responses can enhance object contours but are vulnerable to degradation-induced noise, whereas region-coherent responses improve semantic completeness but may blur object boundaries. Existing methods rarely explicitly consider the spatial variation in structural cue reliability under underwater image degradation. To address this problem, this work proposes SASC-USOD, a novel framework for learning spatially adaptive structural coordination in USOD. The proposed framework constructs two complementary structural representations with different characteristics. A boundary-sensitive representation is obtained by combining fixed Laplacian filtering with a learnable local-detail transformation to enhance discriminative boundary information, while a region-coherent representation is generated through dual-range anisotropic large-kernel contextual aggregation to capture long-range structural consistency. A spatial coordination module is then introduced to estimate the relative reliability of these structural representations and adaptively coordinate their contributions according to image content. Extensive experiments on the USOD10K and USOD benchmarks demonstrate that SASC-USOD consistently outperforms existing methods, reducing MAE by 4.07\% and 23.53\% compared with the strongest competing method, respectively. Moreover, its lightweight variant runs at 21 FPS on an NVIDIA Jetson TX2 NX, demonstrating its capability for onboard underwater robotic perception.

cs.CV↗

arXiv2026

Pandora cluster Lensing, AGN, and Transient Exploration (PLATE) from JWST Multi-Epoch Imaging. I. Discovery of a type II supernova candidate in a spiral galaxy at $z=0.7$

We report the discovery and multi-wavelength analysis of a $z\sim0.7$ transient PLATE-23a in the Abell 2744 field, as the first results of the Pandora Lensing, AGN, and Transient Exploration (PLATE) project. Using multi-epoch JWST NIRCam imaging spanning from 2022 to 2025, we detect PLATE-23a in 12 filters. Difference-imaging analysis reveals its rising and declining phases. The host galaxy of PLATE-23a is a barred spiral at $z=0.688$ with a stellar mass of $\sim 10^{10.4}M_{\odot}$ and a star formation rate of $\sim5.1M_{\odot}~\rm yr^{-1}$, placing it on the star-forming main sequence. Bayesian light-curve classification strongly favors a type IIP supernova (SN) origin. It lies $\sim 13\rm kpc$ (after lensing correction) from the galaxy center in a region of low local star formation, suggesting the progenitor may have migrated from a distant star-forming clump. Physical properties derived from blackbody modeling indicate a temperature decreasing from $\sim8010\rm K$ to $\sim6100\rm K$ at around 65 days after the explosion; the late-time SED is consistent with entering the radioactive decay phase at about one rest-frame year. This work demonstrates the power of deep, multi-epoch JWST observations for studying transients at cosmological distances.

astro-ph.GA↗

arXiv2026

AEGIS: Attention-Embedding Gradient Isolation Shield - Triple-Channel Gradient Masking for Privacy-Preserving Federated LLM Fine-Tuning

Gradient inversion attacks recover private training text from gradients shared in federated learning, posing a serious threat to collaborative model training. Through our analysis of transformer gradient structure, we identify three channels through which private token information leaks: the attention output projection gradient exposes a low-rank subspace that encodes input embeddings (Channel 1), the embedding gradient's row-norm sparsity directly reveals which tokens are present (Channel 2), and the MLP expansion gradient carries a recoverable subspace signal analogous to Channel 1 (Channel 3). State-of-the-art attacks exploit these channels analytically to achieve near-exact token recovery in seconds. Existing defences address at most one channel and either degrade model utility or leave the remaining structural signals intact. We introduce AEGIS (Attention-Embedding Gradient Isolation Shield), a lightweight defence that closes all three analytical channels with three backward-path operations requiring no architectural changes: freezing attention projection parameters eliminates Channel 1 by construction, calibrated noise injection into the embedding gradient destroys Channel 2's token-presence signal, and analogous per-block noise injection into the MLP expansion gradient masks Channel 3. The same masked gradient drives both the local optimiser step and the server export, so no clean signal is retained on either side. Evaluated across 11 models and six datasets, AEGIS reduces token recovery rates to near zero against a range of gradient inversion attacks, both analytical and optimisation-based, while preserving or improving model utility. We provide formal guarantees for Channels 1 and 2 and validate the full defence empirically against adaptive adversaries with complete knowledge of the mechanism.

cs.CR↗

arXiv2026

Universal quantum state purification under energy-preserving constraints

Pure quantum states are a fundamental resource for quantum information processing, but unavoidable noise degrades their purity and limits the performance of quantum tasks. Quantum state purification attempts to restore purity by jointly processing several noisy copies of an unknown state, typically in a probabilistic way. Whether purification is possible, and how much it can achieve, depends strongly on the set of operations that are physically allowed. In this work, we study universal purification under \textit{energy-preserving operations}, which cannot exchange energy with an external environment. We identify exactly when the energy-preserving constraint forbids any universal purification advantage, and we provide concrete examples in which an advantage available to unrestricted operations disappears under energy conservation. When energy-preserving purification remains possible, we determine its optimal fidelity and success probability and construct the corresponding protocols. The framework includes unrestricted purification as a special case and extends naturally to purification assisted by an external energy resource.

quant-ph↗

arXiv2026

Automating Variational Quantum Sensing through Reinforcement-Learned Circuit Structures

Variational quantum sensing offers a promising route to high-precision parameter estimation, but its performance depends strongly on the circuit architectures used for probe preparation and measurement. Existing approaches typically optimize continuous parameters within predefined ansätze, restricting the accessible design space and limiting adaptation to sensing tasks and hardware constraints. Here, we introduce \textsc{AutoQSense}, a reinforcement-learning framework that searches circuit architectures using Fisher-information-based objectives. For few-qubit systems, a single agent sequentially constructs preparation and measurement circuits. For larger systems, a distributed formulation assigns local circuit design to subsystem agents and inter-block entanglement to a budgeted agent. Numerical results show that the learned architectures recover known benchmark strategies, adapt to dephasing noise, and outperform fixed hardware-efficient ansätze while using fewer entangling gates. These results establish \textsc{AutoQSense} as a resource-aware approach to adaptive and hardware-compatible quantum sensing.

quant-ph↗