arXiv subjects

Jing Ren

Publications and source records attributed to Jing Ren.

At least 19 recordsLinked to original sources

arXiv2026

Stage-aware Multimodal Brain Graph Contrastive Learning for Neurodegenerative Disorder Diagnosis

Neurodegenerative disorders are characterized by their progressive nature, with distinct stages that require precise identification and tracking of progression trajectories for accurate diagnosis. However, existing diagnostic methods often fail to fully capture the unique brain patterns associated with different stages, limiting their ability to effectively target specific stages of the disorders. To address this issue, we propose a Stage-aware Multimodal Brain Graph Contrastive Learning (smBGC) method. Specifically, smBGC introduces a stage-aware feature enhancement strategy to capture stage-specific features across all progression stages, enabling the generation of stage-aware multimodal brain graph representations. In addition, smBGC incorporates a stage-adaptive brain graph augmentation approach within a graph contrastive learning framework, facilitating multimodal fusion and improving the model's capability to distinguish between different disorder stages. Experiments on two real-world datasets demonstrate the superior performance of smBGC in various neurodegenerative disorder diagnosis tasks compared with state-of-the-art methods.

cs.AI↗

arXiv2026

Focal-Plane Extreme Adaptive Optics for Deep Super Resolution Imaging and Nanolithography

Super resolution optical imaging and nanolithography enable nanoscale observation and fabrication, but high order, spatially varying phase and amplitude aberrations in optically heterogeneous media severely limit their performance at depth. Here we introduce focal-plane extreme adaptive optics (FP-ExAO), which moves beyond conventional phase conjugate correction by directly sensing and controlling the focal-plane complex field. This terminal field captures aberrations accumulated throughout the optical system. A state space formulation directly links deformable mirror commands to the focal field, avoiding explicit plane by plane Fourier optics modelling, while information theoretically optimal probing and closed-form control enable rapid field estimation and correction. Cascaded deformable mirrors further enable FP-ExAO to correct coupled phase-amplitude and spatially varying aberrations. In three-dimensional (3D) single molecule localization microscopy (SMLM), FP-ExAO enables imaging to 160 um depth in brain tissue and beyond 60 um in kidney tissue, while maintaining approximately 18 nm localization precision across a 75 um x 75 um corrected field, with a 16-fold larger field of view and more than tenfold faster correction than conventional AO methods. These capabilities meet the imaging requirements for 3D deep tissue optical connectome reconstruction and quantitative super resolution pathology. In stimulated emission depletion (STED) nanolithography, FP-ExAO reduces the minimum written linewidth from 251.1 nm without correction to 88.8 nm at a depth of 160 um in photoresist. FP-ExAO establishes a new adaptive optics paradigm for deep nanoscale optical readout and writing.

physics.optics↗

arXiv2026

Why Are LLMs Confidently Wrong? Correcting Overconfident Errors via Causal Head Intervention

Modern large language models (LLMs) often exhibit overconfident errors, assigning high confidence to incorrect predictions and thereby undermining reliability in real-world use. Existing calibration methods typically rely on post-hoc adjustment or retraining, operating at the output level without addressing the underlying causes of miscalibration. We propose CoCHI, (Class-oriented Causal Head Intervention), a framework that improves calibration by identifying and intervening on internal components responsible for distinct confidence behaviours. By analysing predictions through correctness-confidence patterns, CoCHI enables targeted intervention on model internals, reducing overconfident errors while preserving correct predictions. Our approach provides a mechanistic perspective on calibration, linking reliability to identifiable structures within the model rather than treating it as an output-level issue. Experiments across multiple LLMs show consistent improvements in calibration metrics, including Expected Calibration Error, Negative Log-Likelihood, and Brier Score, while maintaining accuracy on five multiple-choice question answering benchmarks. These results suggest that effective calibration requires not only adjusting outputs, but also addressing the underlying causal mechanisms within the model.

cs.AI↗

arXiv2026

Causal Green function decomposition for quantum black hole seismology

The growing sensitivity of gravitational-wave detectors enables increasingly precise tests of black hole (BH) ringdown spectroscopy. BH quasinormal modes (QNMs) are, however, spectrally unstable: small near-horizon modifications can produce a drastically different QNM spectrum, while causality requires the prompt ringdown to remain BH-like until the reflected signal returns. Quantum BHs with substantial interior reflection provide a natural setting for this tension, yet the relation between their time-domain waveform and different QNM spectra still lacks a consistent picture. In this work, we systematically examine the time-domain Green function for quantum BHs, considering sources located outside and inside the light-ring potential barrier. By decomposing the Green function into causally distinct components and choosing the corresponding inverse-Laplace contours consistently, we clarify how the response is built from different sets of QNMs. We find that the quantum BH QNM reconstruction always faithfully describes the signal once the curved spacetime is probed, but its practical efficiency depends strongly on the evolutionary stage. Before interior reflection becomes relevant, we prove that this basis is formally equivalent to the BH QNM and tail expansions, with convergence properties sensitive to source location. At late times, the long-lived modes provide an efficient basis. Time-domain simulations confirm these results, providing a unified causal framework for BH spectroscopy and quantum BH seismology.

gr-qc↗

arXiv2026

Model-agnostic search of gravitational wave echoes in LVK data

Gravitational wave echoes offer a unique probe of the near-horizon structure of astrophysical black holes, beyond the standard "black hole spectroscopy." Theoretical waveform predictions, however, remain uncertain, motivating robust searches that avoid specific echo modeling. We present a model-agnostic search framework targeting long-lived quasinormal modes (QNMs) expected from strong interior reflection. By employing a generalized phase-marginalized likelihood that coherently combines data for each QNM across a detector network, our method enhances sensitivity to the signals. To handle real detector noise, we implement an optimized notching procedure to suppress instrumental spectral lines and refine the Bayesian parameter settings. We validate the performance of this framework using injection studies on O1 background data, demonstrating reliable signal recovery in realistic noise conditions. We then apply this method to three binary black hole merger events with high ringdown signal-to-noise ratios (SNRs): GW150914 from O1, GW231226 from O4a, and the recently reported O4 event GW250114. No statistically significant evidence for postmerger echoes is found. Consequently, we derive 90% upper limits on the network SNR and the average initial strain amplitude of the long-lived QNMs. These results provide model-agnostic constraints on late-time echoes from LVK data, complementing existing searches for other echo signatures.

gr-qc↗

arXiv2026

The MeerKAT Thousand-Pulsar Polarisation Array II: Searches for Ultralight Axion-Like Dark Matter

We construct Pulsar Polarisation Arrays (PPA), using regular pulsars monitored in MeerKAT's Thousand Pulsar Array (TPA) Programme, to search for Axion-like Dark Matter (ALDM) within Milky Way. Specifically, from a catalogue of 1237 regular pulsars, we select the 50 ones with the highest signal-to-noise ratio and set upper limits on the ALDM Chern-Simons coupling. We find no signals with statistical significance over the mass range of $[10^{-23},10^{-20}]\,{\rm eV}$ in the six-year MeerKAT's data. By combining the high-quality TPA pulsars and the accurate ionospheric subtraction of spinifex, we establish the most sensitive upper limits to the date on the ALDM Chern-Simons coupling, namely $\lesssim 10^{-14} - 3\times 10^{-13}\,{\rm GeV}^{-1}$, for the mass range of $[10^{-23},10^{-21}]\,{\rm eV}$ except at $m_a \sim 1.3 \times 10^{-22}\,$eV. This study underscores the great potential of constructing regular-pulsar PPAs for scientific tasks.

astro-ph.HE↗

arXiv2026

The MeerKAT Thousand-Pulsar Polarization Array I: Properties of the Polarization and Rotation Measure Time Series Data

The polarimetry of recent pulsar observations has provided a wealth of observational data with which to test physical theories of emission mechanisms, radiative transfer and even theories that extend beyond the Standard Model. In this work, we have outlined the data analysis of the polarisation time series data of a population of 513 pulsars from the Thousand Pulsar Array observing programme, laying the foundation for building the MeerKAT Thousand-Pulsar Polarization Array as a probe for ultralight Axion-Like Dark Matter (ALDM). From this large dataset we have focused on the temporal trends in the observed polarisation angle (PA) through a measure we call the PA offset, and characterised the trends due to the effects of Faraday Rotation within the interstellar medium and the Earth's ionosphere, as well as generic white and red noise models that are estimated within a Bayesian MCMC analysis. Then, motivated by potential extra contributions to the rotation of the PA that may not be Faraday-like, arising from the proposed ALDM field, we have investigated a derived time dependence for the rotation measure (RM) required to explain the observed PA offset. Comparison of these estimates to RM values that are measured in typical pulsar studies, through a technique known as RM Synthesis, provides a probe of any wavelength-independent contribution to the rotation of the PA. Although we find no evidence for oscillatory behaviour within our dataset's observation timespan, we do find cases of deviation from the usual RM values in several `pulsars of interest', as well as long-term linear trends in the time evolution of Faraday rotation that have been presented in the literature before.

astro-ph.HE↗

arXiv2026

Impact of Satellite Constellations on Observations with the 80-cm Telescope and the Mini-SiTian at the Xinglong Observatory, NAOC

The rapid development of mega-constellations in low Earth orbit (LEO) severely impacts ground-based optical astronomical observations. By combining WorldWide Telescope (WWT) simulations with 2019 and 2023 observational data from the Xinglong Observatory 80-cm telescope and 2023 data from the Mini-SiTian (MST), we find that satellite visibility increases with deployment, particularly during the summer. For the 80-cm telescope, the fraction of images containing satellite trails increased from an average of 0.34% in 2019 to 0.7% in 2023; meanwhile, for the MST in 2023, the fraction rose from 5% in January to 12% by December, peaking at 19% in the summer. Through stratified analysis of solar elevation and local time, we find that observations during twilight and summer are particularly susceptible to satellite trail interference. Photometric analysis reveals that the interference intensity increases for fainter sources and those closer to the trails. Furthermore, a comparative analysis across different seeing conditions shows that the deviation of median standardized residuals (σ) is significantly greater under poor seeing than under good seeing conditions.

astro-ph.IM↗

arXiv2026

Two-Layer Reinforcement Learning-Assisted Joint Beamforming and Trajectory Optimization for Multi-UAV Downlink Communications

Unmanned aerial vehicles (UAVs) are pivotal for future 6G non-terrestrial networks, yet their high mobility creates a complex coupled optimization problem for beamforming and trajectory design. Existing numerical methods suffer from prohibitive latency, while standard deep learning often ignores dynamic interference topology, limiting scalability. To address these issues, this paper proposes a hierarchically decoupled framework synergizing graph neural networks (GNNs) with multi-agent reinforcement learning. Specifically, on the short timescale, we develop a topology-aware GNN beamformer by incorporating GraphNorm. By modeling the dynamic UAV-user association as a time-varying heterogeneous graph, this method explicitly extracts interference patterns to achieve sub-millisecond inference. On the long timescale, trajectory planning is modeled as a decentralized partially observable Markov decision process and solved via the multi-agent proximal policy optimization algorithm under the centralized training with decentralized execution paradigm, facilitating cooperative behaviors. Extensive simulation results demonstrate that the proposed framework significantly outperforms state-of-the-art optimization heuristics and deep learning baselines in terms of system sum rate, convergence speed, and generalization capability.

eess.SP↗

arXiv2026

Bottleneck Tokens for Unified Multimodal Retrieval

Adapting decoder-only multimodal large language models (MLLMs) for unified multimodal retrieval faces two structural gaps. First, existing methods rely on implicit pooling, which overloads the hidden state of a standard vocabulary token (e.g., ) as the sequence-level representation, a mechanism never designed for information aggregation. Second, contrastive fine-tuning specifies what the embedding should match but provides no token-level guidance on how information should be compressed into it. We address both gaps with two complementary components. Architecturally, we introduce Bottleneck Tokens (BToks), a small set of learnable tokens that serve as a fixed-capacity explicit pooling mechanism. For training, we propose Generative Information Condensation: a next-token prediction objective coupled with a Condensation Mask that severs the direct attention path from target tokens to query tokens. All predictive signals are thereby forced through the BToks, converting the generative loss into dense, token-level supervision for semantic compression. At inference time, only the input and BToks are processed in a single forward pass with negligible overhead over conventional last-token pooling. On MMEB-V2 (78 datasets, 3 modalities, 9 meta-tasks), our approach achieves state-of-the-art among 2B-scale methods under comparable data conditions, attaining an Overall score of 59.0 (+3.6 over VLM2Vec-V2) with substantial gains on semantically demanding tasks (e.g., +12.6 on Video-QA).

cs.LG↗

arXiv2026

Reinforcement Learning-Guided NSGA-II Enhanced with Gray Relational Coefficient for Multi-Objective Optimization: Application to NASDAQ Portfolio Optimization

In modern financial markets, decision-makers increasingly rely on quantitative methods to navigate complex trade-offs among multiple, often conflicting objectives. This paper addresses constrained multi-objective optimization (MOO) with an application to portfolio optimization for minimizing risk and maximizing return. To address existing gaps, we propose a novel reinforcement learning (RL)-guided non-dominated sorting genetic algorithm II (NSGA-II) enhanced with gray relational coefficients (GRC), termed RL-NSGA-II-GRC, which combines an RL agent controller and GRC-based selection to improve convergence and diversity of Pareto fronts. The agent adapts evolutionary parameters online using metrics of hypervolume, feasibility, and diversity, while the GRC tournament operator ranks parents via a unified score considering dominance rank, crowding distance, and proximity to ideal reference. We evaluate the framework on the Kursawe and CONSTR benchmarks and a NASDAQ portfolio application. On the benchmarks, RL-NSGA-II-GRC achieves convergence improvements of about 5.8% and 4.4% over NSGA-II, while preserving well-distributed non-dominated solutions. In the portfolio application, it produces a smooth, densely populated efficient frontier supporting identification of the maximum Sharpe ratio portfolio (annualized Sharpe =1.92) and utility-optimal portfolios for different risk-aversion levels. The main contributions are three-fold: 1) we propose an RL-NSGA-II-GRC method integrating an RL agent into the evolutionary framework to adaptively control parameters via generational feedback; 2) we design a GRC-enhanced binary tournament operator providing a comprehensive indicator to guide the search toward the Pareto front; 3) we demonstrate, on benchmark MOO and a NASDAQ case study, that the method delivers improved convergence and well-populated frontiers supporting actionable insights.

cs.LG↗

arXiv2026

Probing Ultralight Axion-like Dark Matter: A Pulsar Timing Arrays-Pulsar Polarization Arrays Synergy

Ultralight axionlike dark matter (ALDM) is a leading candidate in the dark matter realm, characterized by its prominent wave properties on astronomical scales. Pulsar timing arrays (PTAs) and Pulsar polarization arrays (PPAs) aim to detect this dark matter through timing and polarization measurements, respectively, of pulsars. The PTA relies on gravitational effects, while the PPA detects nongravitational effects. These two methods complement each other, synergistically enhancing the pulsar array's capability to identify the ALDM signals in the data. In this article, we provide a foundational development of this synergy. We begin by revisiting previously derived two-point correlation functions for both PTA and PPA, and extend the analysis to include the leading-order correlation between timing and polarization signals, encoded as a three-point function. We then explore the constructions of likelihood functions for PTA and combined PTA-PPA analyses within a Bayesian framework, aimed at detecting the characteristic correlations of ALDM signals. We emphasize the non-Gaussianity of the ALDM timing signals, which arises from their nonlinear dependence on the field, in contrast to the Gaussian nature of its polarization signals. To address the complexities introduced, we approach this investigation in two ways: one involves a Gaussian approximation with proper justifications, while the other derives the formalism from the generic Gaussian characteristics of the ALDM field. Particularly, for the combined PTA-PPA case, a clear connection between the three-point correlation function and the likelihood is established. We anticipate that these efforts will lead to further developments in PTA and PTA-PPA analysis methods, better accounting for the influence of non-Gaussianity.

astro-ph.CO↗

arXiv2026

Geometry Distributions

Neural representations of 3D data have been widely adopted across various applications, particularly in recent work leveraging coordinate-based networks to model scalar or vector fields. However, these approaches face inherent challenges, such as handling thin structures and non-watertight geometries, which limit their flexibility and accuracy. In contrast, we propose a novel geometric data representation that models geometry as distributions-a powerful representation that makes no assumptions about surface genus, connectivity, or boundary conditions. Our approach uses diffusion models with a novel network architecture to learn surface point distributions, capturing fine-grained geometric details. We evaluate our representation qualitatively and quantitatively across various object types, demonstrating its effectiveness in achieving high geometric fidelity. Additionally, we explore applications using our representation, such as textured mesh representation, neural surface compression, dynamic object modeling, and rendering, highlighting its potential to advance 3D geometric learning.

cs.CV↗

arXiv2026

Spiking Graph Predictive Coding for Reliable OOD Generalization

Graphs provide a powerful basis for modeling Web-based relational data, with expressive GNNs to support the effective learning in dynamic web environments. However, real-world deployment is hindered by pervasive out-of-distribution (OOD) shifts, where evolving user activity and changing content semantics alter feature distributions and labeling criteria. These shifts often lead to unstable or overconfident predictions, undermining the trustworthiness required for Web4Good applications. Achieving reliable OOD generalization demands principled and interpretable uncertainty estimation; however, existing methods are largely post-hoc, insensitive to distribution shifts, and unable to explain where uncertainty arises especially in high-stakes settings. To address these limitations, we introduce SpIking GrapH predicTive coding (SIGHT), an uncertainty-aware plug-in graph learning module for reliable OOD Generalization. SIGHT performs iterative, error-driven correction over spiking graph states, enabling models to expose internal mismatch signals that reveal where predictions become unreliable. Across multiple graph benchmarks and diverse OOD scenarios, SIGHT consistently enhances predictive accuracy, uncertainty estimation, and interpretability when integrated with GNNs.

cs.LG↗

arXiv2026

Probing Boson Clouds with Supermassive Black Hole Binaries

Rotating black holes can generate boson clouds via superradiance when the boson's Compton wavelength is comparable to the black hole's size. In binary systems, these clouds can produce distinctive observational imprints. Recent studies accounting for nonlinearities induced by orbital backreaction suggest that if the binary forms at a large separation, resonance transitions can significantly deplete the cloud, minimizing later observational consequences except for very specific orbital inclinations. In this paper, we extend this framework to supermassive black hole binaries (SMBHBs), considering the influence of their astrophysical evolutionary histories. We find that, before entering the gravitational wave (GW) radiation stage, the additional energy loss channels can accelerate orbital evolution. This acceleration makes hyperfine resonant transitions inefficient, allowing a sufficient portion of the cloud to remain for later direct observations. We further discuss the ionization effects and cloud depletion occurring at this stage. Based on these theoretical insights, we explore how multi-messenger observations for SMBHBs can be utilized to detect the ionization effects of boson clouds by examining changes in the orbital period decay rate via electromagnetic measurements and variations in GW strain over a wide frequency band. Our findings reveal a complex dependence on the binary's total mass, mass ratio, and boson mass, emphasizing the significant role of astrophysical evolution histories in detecting boson clouds within binaries.

hep-ph↗

arXiv2026

When to Trust: A Causality-Aware Calibration Framework for Accurate Knowledge Graph Retrieval-Augmented Generation

Knowledge Graph Retrieval-Augmented Generation (KG-RAG) extends the RAG paradigm by incorporating structured knowledge from knowledge graphs, enabling Large Language Models (LLMs) to perform more precise and explainable reasoning. While KG-RAG improves factual accuracy in complex tasks, existing KG-RAG models are often severely overconfident, producing high-confidence predictions even when retrieved sub-graphs are incomplete or unreliable, which raises concerns for deployment in high-stakes domains. To address this issue, we propose Ca2KG, a Causality-aware Calibration framework for KG-RAG. Ca2KG integrates counterfactual prompting, which exposes retrieval-dependent uncertainties in knowledge quality and reasoning reliability, with a panel-based re-scoring mechanism that stabilises predictions across interventions. Extensive experiments on two complex QA datasets demonstrate that Ca2KG consistently improves calibration while maintaining or even enhancing predictive accuracy.

cs.CL↗

arXiv2026

FairGE: Fairness-Aware Graph Encoding in Incomplete Social Networks

Graph Transformers (GTs) are increasingly applied to social network analysis, yet their deployment is often constrained by fairness concerns. This issue is particularly critical in incomplete social networks, where sensitive attributes are frequently missing due to privacy and ethical restrictions. Existing solutions commonly generate these incomplete attributes, which may introduce additional biases and further compromise user privacy. To address this challenge, FairGE (Fair Graph Encoding) is introduced as a fairness-aware framework for GTs in incomplete social networks. Instead of generating sensitive attributes, FairGE encodes fairness directly through spectral graph theory. By leveraging the principal eigenvector to represent structural information and padding incomplete sensitive attributes with zeros to maintain independence, FairGE ensures fairness without data reconstruction. Theoretical analysis demonstrates that the method suppresses the influence of non-principal spectral components, thereby enhancing fairness. Extensive experiments on seven real-world social network datasets confirm that FairGE achieves at least a 16% improvement in both statistical parity and equality of opportunity compared with state-of-the-art baselines. The source code is shown in https://github.com/LuoRenqiang/FairGE.

cs.SI↗

arXiv2026

When to Invoke: Refining LLM Fairness with Toxicity Assessment

Large Language Models (LLMs) are increasingly used for toxicity assessment in online moderation systems, where fairness across demographic groups is essential for equitable treatment. However, LLMs often produce inconsistent toxicity judgements for subtle expressions, particularly those involving implicit hate speech, revealing underlying biases that are difficult to correct through standard training. This raises a key question that existing approaches often overlook: when should corrective mechanisms be invoked to ensure fair and reliable assessments? To address this, we propose FairToT, an inference-time framework that enhances LLM fairness through prompt-guided toxicity assessment. FairToT identifies cases where demographic-related variation is likely to occur and determines when additional assessment should be applied. In addition, we introduce two interpretable fairness indicators that detect such cases and improve inference consistency without modifying model parameters. Experiments on benchmark datasets show that FairToT reduces group-level disparities while maintaining stable and reliable toxicity predictions, demonstrating that inference-time refinement offers an effective and practical approach for fairness improvement in LLM-based toxicity assessment systems. The source code can be found at https://aisuko.github.io/fair-tot/.

cs.CL↗