arXiv subjects

Qin Li

Publications and source records attributed to Qin Li.

At least 73 records · Page 4Linked to original sources

arXiv2025

D-SMART: Enhancing LLM Dialogue Consistency via Dynamic Structured Memory And Reasoning Tree

Large Language Models (LLMs) often exhibit factual inconsistencies and logical decay in extended, multi-turn dialogues, a challenge stemming from their reliance on static, pre-trained knowledge and an inability to reason adaptively over the dialogue history. Prevailing mitigation strategies, such as Retrieval-Augmented Generation (RAG) and agentic working memories, improve information recall but still engage with fundamentally static knowledge sources and follow pre-defined single reasoning path. This hinders their ability to preserve factual and logical consistency of their responses in multi-turn dialogues while the context evolves over time. To address this issue, we propose D-SMART, a model-agnostic framework designed to maintain multi-turn dialogue consistency by enabling LLMs to build and reason over a dynamic, structured representation of the conversational context. This is achieved via two synergistic components: (1) a Dynamic Structured Memory (DSM), which incrementally constructs and maintains an authoritative, OWL-compliant knowledge graph of the conversation; and (2) a Reasoning Tree (RT), which executes inferences as an explicit and traceable multi-step search over the graph. As the popular-used quality score (judged by GPT-4) can overlook logical flaws, we introduce new NLI-based metrics to better measure multi-turn dialogue consistency. Comprehensive experiments on the MT-Bench-101 benchmark show that D-SMART significantly outperforms state-of-the-art baselines, elevating the dialogue consistency score by over 48\% for both proprietary and open-source models, and notably improves the quality score of the latter by up to 10.1\%.

cs.CL↗

arXiv2025

Physics-informed Attention-enhanced Fourier Neural Operator for Solar Magnetic Field Extrapolations

We propose Physics-informed Attention-enhanced Fourier Neural Operator (PIANO) to solve the Nonlinear Force-Free Field (NLFFF) problem in solar physics. Unlike conventional approaches that rely on iterative numerical methods, our proposed PIANO directly learns the 3D magnetic field structure from 2D boundary conditions. Specifically, PIANO integrates Efficient Channel Attention (ECA) mechanisms with Dilated Convolutions (DC), which enhances the model's ability to capture multimodal input by prioritizing critical channels relevant to the magnetic field's variations. Furthermore, we apply physics-informed loss by enforcing the force-free and divergence-free conditions in the training process so that our prediction is consistent with underlying physics with high accuracy. Experimental results on the ISEE NLFFF dataset show that our PIANO not only outperforms state-of-the-art neural operators in terms of accuracy but also shows strong consistency with the physical characteristics of NLFFF data across magnetic fields reconstructed from various solar active regions. The GitHub of this project is available https://github.com/Autumnstar-cjh/PIANO

cs.LG↗

arXiv2025

Comparing LSTM-Based Sequence-to-Sequence Forecasting Strategies for 24-Hour Solar Proton Flux Profiles Using GOES Data

Solar Proton Events (SPEs) cause significant radiation hazards to satellites, astronauts, and technological systems. Accurate forecasting of their proton flux time profiles is crucial for early warnings and mitigation. This paper explores deep learning sequence-to-sequence (seq2seq) models based on Long Short-Term Memory networks to predict 24-hour proton flux profiles following SPE onsets. We used a dataset of 40 well-connected SPEs (1997-2017) observed by NOAA GOES, each associated with a >=M-class western-hemisphere solar flare and undisturbed proton flux profiles. Using 4-fold stratified cross-validation, we evaluate seq2seq model configurations (varying hidden units and embedding dimensions) under multiple forecasting scenarios: (i) proton-only input vs. combined proton+X-ray input, (ii) original flux data vs. trend-smoothed data, and (iii) autoregressive vs. one-shot forecasting. Our major results are as follows: First, one-shot forecasting consistently yields lower error than autoregressive prediction, avoiding the error accumulation seen in iterative approaches. Second, on the original data, proton-only models outperform proton+X-ray models. However, with trend-smoothed data, this gap narrows or reverses in proton+X-ray models. Third, trend-smoothing significantly enhances the performance of proton+X-ray models by mitigating fluctuations in the X-ray channel. Fourth, while models trained on trendsmoothed data perform best on average, the best-performing model was trained on original data, suggesting that architectural choices can sometimes outweigh the benefits of data preprocessing.

cs.LG↗

arXiv2025

Improving the Spatial Resolution of GONG Solar Images to GST Quality Using Deep Learning

High-resolution (HR) solar imaging is crucial for capturing fine-scale dynamic features such as filaments and fibrils. However, the spatial resolution of the full-disk H$α$ images is limited and insufficient to resolve these small-scale structures. To address this, we propose a GAN-based superresolution approach to enhance low-resolution (LR) full-disk H$α$ images from the Global Oscillation Network Group (GONG) to a quality comparable with HR observations from the Big Bear Solar Observatory/Goode Solar Telescope (BBSO/GST). We employ Real-ESRGAN with Residual-in-Residual Dense Blocks and a relativistic discriminator. We carefully aligned GONG-GST pairs. The model effectively recovers fine details within sunspot penumbrae and resolves fine details in filaments and fibrils, achieving an average mean squared error (MSE) of 467.15, root mean squared error (RMSE) of 21.59, and cross-correlation (CC) of 0.7794. Slight misalignments between image pairs limit quantitative performance, which we plan to address in future work alongside dataset expansion to further improve reconstruction quality.

cs.CV↗

arXiv2025

Consistent Assistant Domains Transformer for Source-free Domain Adaptation

Source-free domain adaptation (SFDA) aims to address the challenge of adapting to a target domain without accessing the source domain directly. However, due to the inaccessibility of source domain data, deterministic invariable features cannot be obtained. Current mainstream methods primarily focus on evaluating invariant features in the target domain that closely resemble those in the source domain, subsequently aligning the target domain with the source domain. However, these methods are susceptible to hard samples and influenced by domain bias. In this paper, we propose a Consistent Assistant Domains Transformer for SFDA, abbreviated as CADTrans, which solves the issue by constructing invariable feature representations of domain consistency. Concretely, we develop an assistant domain module for CADTrans to obtain diversified representations from the intermediate aggregated global attentions, which addresses the limitation of existing methods in adequately representing diversity. Based on assistant and target domains, invariable feature representations are obtained by multiple consistent strategies, which can be used to distinguish easy and hard samples. Finally, to align the hard samples to the corresponding easy samples, we construct a conditional multi-kernel max mean discrepancy (CMK-MMD) strategy to distinguish between samples of the same category and those of different categories. Extensive experiments are conducted on various benchmarks such as Office-31, Office-Home, VISDA-C, and DomainNet-126, proving the significant performance improvements achieved by our proposed approaches. Code is available at https://github.com/RoryShao/CADTrans.git.

cs.CV↗

arXiv2025

Evaporative Refrigeration Effect in Evaporation and Condensation between Two Parallel Plates

It is well-known that evaporation can lead to cooling. However, little is known that evaporation can actually create a refrigeration effect, i.e., the vapor phase temperature can drop below the temperature of the cooling wall. This possibility was recently pointed out via modeling based on an approximate quasi-continuum approach. This work examines this effect rigorously by studying evaporation and condensation between two parallel plates by coupling the solution of the Boltzmann transport equation in the vapor phase with the continuum treatments in both liquid films. Numerical results show that the vapor phase temperature at the evaporating side can be much lower than the coldest wall temperature at the condensing surface, reaffirming the evaporative refrigeration effect. The present work further reveals that this effect is caused by two mechanisms. While the dominant mechanism is the asymmetry in the molecular distribution between the outgoing and the incoming molecules at the interface, additional cooling occurs within the Knudsen layer due to the sudden expansion, similar to the Joule-Thomson effect, although with subtle differences in that the interfacial expansion is not an isenthalpic process. The impacts of key parameters, including liquid thickness, Knudsen number, and accommodation coefficient, are investigated. The numerical simulation shows that with a thicker vapor, a thinner liquid film, and a larger accommodation coefficient, leads to stronger evaporative refrigeration effect. This work will motivate future experiments to further confirm this prediction and explore its potential applications in air-conditioning, refrigeration, and membrane distillation.

cond-mat.stat-mech↗

arXiv2025

Control of kinetic plasma instabilities by laser fields

We study the possibility of controlling kinetic plasma instabilities by using lasers to apply external electromagnetic fields. We derive the dispersion relation for the corresponding mathematical description, a reduced Vlasov--Maxwell system, by extending the well-known Penrose condition. It is observed that, under very mild assumptions, the dispersion relation decouples into two parts. The first part is identical to the classic Penrose condition for the Vlasov--Poisson system, while the second part describes the influence of the laser on the transverse dynamics (e.g. a Weibel instability). In particular, this means that the longitudinal dynamics (e.g. a two-stream instability) can not be stabilized in this manner as far as linear theory is concerned. We show, however, that nonlinear effects can be used to couple the two parts and achieve effective control. This is done by determining the control parameters (i.e. the form of the external electric and magnetic fields) by solving a PDE-constrained optimization problem.

physics.plasm-ph↗

arXiv2025

Numerical study of friction reduction and underlying mechanism of air film via parallel and tandem injections on a hypersonic flat plate at high altitude

This study uses numerical simulations to analyze a flat-plate with air films generated via parallel and tandem injections under freestream conditions of Ma=15 at an altitude of H=60km. Simulations are carried out using various mass flow rates and jet spacing, followed by an examination of the effect of the streamwise-to-spanwise aspect ratio of the injection hole. The following key findings are acquired: (1) Under hypersonic and high-altitude conditions, the air injection lifts the incoming flow, producing a pronounced low-speed region near the injection site and over the downstream wall. A plateau appears in the wall-normal distribution of the streamwise velocity, with a value significantly lower than that of its freestream counterpart. This reduces the near-wall velocity gradient, thereby achieving drag reduction. However, the subsequent flow reattachment may lead to localized drag increases; (2) As the mass flow rate of injection increases, the wall-normal extent of the near-wall low-speed layer expands and the film coverage broadens; however, the increased friction limits gains in drag reduction, and greater flow losses are also incurred; (3) Increasing the hole spacing in the parallel injection reduces the penetration height of the film, the spanwise coverage, and the drag reduction efficiency, as well as decreasing the flow losses. In contrast, for the studied tandem injections, no significant differences are observed with analogous spacing variations; (4) Increasing the injection aspect ratio enlarges the spanwise coverage of the low-speed layer, improves the spanwise uniformity of the streamwise velocity distribution, diminishes the friction-increasing zones, and significantly enhances the drag reduction. Notably, the consequences are the inverse of those of the low-speed situation.

physics.flu-dyn↗

arXiv2025

Accuracy analysis and optimization of scale-independent third-order WENO-Z scheme with critical-point accuracy preservation

To address the order degradation at critical points in the WENO3-Z scheme, some improvements have been proposed , but these approaches generally fail to consider the occurrence of critical points at arbitrary positions within grid intervals, resulting in their inability to maintain third-order accuracy when a first-order critical point (CP1) occurs. Also, most previous improved schemes suffer from a relatively large exponent p of the ratio of global to local smoothness indicators, which adversely affects the numerical resolution. Concerning these limitations, introduced here is an accuracy-optimization lemma demonstrating that the accuracy of nonlinear weights can be enhanced providing that smoothness indicators satisfy specific conditions, thereby establishing a methodology for elevating the accuracy of nonlinear weights. Leveraging this lemma, a local smoothness indicator is constructed with error terms achieving second-order in smooth regions and fourth-order at CP1, alongside a global smoothness indicator yielding fourth-order accuracy in smooth regions and fifth-order at CP1, enabling the derivation of new nonlinear weights that meet accuracy requirements even when employing p=1. Furthermore, a resolution-optimization lemma is proposed to analyze the relationship between parameters in local smoothness indicators and resolution. By integrating theoretical analysis with numerical practices, free parameters in non-normalized weights and local smoothness indicators are determined under the balance of numerical resolution and robustness, which leads to the development of WENO3-ZES4, a new WENO3-Z improvement that preserves the optimal order at CP1 especially with p=1. 1D and 2D validating tests show that the new scheme consistently achieves third-order in the case of CP1 regardless of its position and exhibits good resolution as well as preferable robustness.

math.NA↗

arXiv2025

Reconstruction of Solar EUV Irradiance Using CaII K Images and SOHO/SEM Data with Bayesian Deep Learning and Uncertainty Quantification

Solar extreme ultraviolet (EUV) irradiance plays a crucial role in heating the Earth's ionosphere, thermosphere, and mesosphere, affecting atmospheric dynamics over varying time scales. Although significant effort has been spent studying short-term EUV variations from solar transient events, there is little work to explore the long-term evolution of the EUV flux over multiple solar cycles. Continuous EUV flux measurements have only been available since 1995, leaving significant gaps in earlier data. In this study, we propose a Bayesian deep learning model, named SEMNet, to fill the gaps. We validate our approach by applying SEMNet to construct SOHO/SEM EUV flux measurements in the period between 1998 and 2014 using CaII K images from the Precision Solar Photometric Telescope. We then extend SEMNet through transfer learning to reconstruct solar EUV irradiance in the period between 1950 and 1960 using CaII K images from the Kodaikanal Solar Observatory. Experimental results show that SEMNet provides reliable predictions along with uncertainty bounds, demonstrating the feasibility of CaII K images as a robust proxy for long-term EUV fluxes. These findings contribute to a better understanding of solar influences on Earth's climate over extended periods.

astro-ph.SR↗

arXiv2025

Ultra3D: Efficient and High-Fidelity 3D Generation with Part Attention

Recent advances in sparse voxel representations have significantly improved the quality of 3D content generation, enabling high-resolution modeling with fine-grained geometry. However, existing frameworks suffer from severe computational inefficiencies due to the quadratic complexity of attention mechanisms in their two-stage diffusion pipelines. In this work, we propose Ultra3D, an efficient 3D generation framework that significantly accelerates sparse voxel modeling without compromising quality. Our method leverages the compact VecSet representation to efficiently generate a coarse object layout in the first stage, reducing token count and accelerating voxel coordinate prediction. To refine per-voxel latent features in the second stage, we introduce Part Attention, a geometry-aware localized attention mechanism that restricts attention computation within semantically consistent part regions. This design preserves structural continuity while avoiding unnecessary global attention, achieving up to 6.7x speed-up in latent generation. To support this mechanism, we construct a scalable part annotation pipeline that converts raw meshes into part-labeled sparse voxels. Extensive experiments demonstrate that Ultra3D supports high-resolution 3D generation at 1024 resolution and achieves state-of-the-art performance in both visual fidelity and user preference.

cs.CV↗

arXiv2025

Solar Alfvenic Pulses and Mesoscale Solar Wind

Large-scale solar ejections are well understood, but the extent to which small-scale solar features directly influence the solar wind remains an open question, primarily due to the challenges of tracing these small-scale ejections and their impact. Here, we measure the fine-scale motions of network bright points along a coronal hole boundary in high-resolution H-alpha images from the 1.6m Goode Solar Telescope at Big Bear Solar Observatory to quantify the agitation of open flux tubes into generating Alfvenic pulses. We combine the motion, magnetic flux, and activity duration of the flux tubes to estimate the energy content carried by individual Alfvenic pulses, which is ~10+25 erg, adequately higher than the energies ~10+23 erg estimated for the magnetic switchbacks observed by the Parker Solar Probe (PSP). This implies the possibility that the surface-generated Alfvenic pulses could reach the solar wind with sufficient energy to generate switchbacks, even though some of then are expected to be reflected back in the stratified solar atmosphere. Alfvenic pulses further reproduce for the first time other properties of switchbacks, including the filling factor above ~8% at granular and supergranular scales, which correspond best to the lower end of the mesoscale structure. This quantitative result for solar energy output in the form of Alfvenic pulses through magnetic funnels provides a crucial clue to the ongoing debate about the dynamic cycle of energy exchange between the Sun and the mesoscale solar wind that has been raised, but has not been adequately addressed, by PSP near-Sun observations.

astro-ph.SR↗

arXiv2025

MVPinn: Integrating Milne-Eddington Inversion with Physics-Informed Neural Networks for GST/NIRIS Observations

We introduce MVPinn, a Physics-Informed Neural Network (PINN) approach tailored for solving the Milne-Eddington (ME) inversion problem, specifically applied to spectropolarimetric observations from the Big Bear Solar Observatory's Near-InfraRed Imaging Spectropolarimeter (BBSO/NIRIS) at the Fe I 1.56 μm lines. Traditional ME inversion methods, though widely used, are computationally intensive, sensitive to noise, and often struggle to accurately capture complex profile asymmetries resulting from gradients in magnetic field strength, orientation, and line-of-sight velocities. By embedding the ME radiative transfer equations directly into the neural network training as physics-informed constraints, our MVPinn method robustly and efficiently retrieves magnetic field parameters, significantly outperforming traditional inversion methods in accuracy, noise resilience, and the ability to handle asymmetric and weak polarization signals. After training, MVPinn infers one magnetogram in about 15 seconds, compared to tens of minutes required by traditional ME inversion on high-resolution spectropolarimetric data. Quantitative comparisons demonstrate excellent agreement with well-established magnetic field measurements from the SDO/HMI and Hinode/SOT-SP instruments, with correlation coefficients of approximately 90%. In particular, MVPINN aligns better with Hinode/SOT-SP data, indicating some saturation of HMI data at high magnetic strengths. We further analyze the physical significance of profile asymmetries and the limitations inherent in the ME model assumption. Our results illustrate the potential of physics-informed machine learning methods in high-spatial-temporal solar observations, preparing for more sophisticated, real-time magnetic field analysis essential for current and next-generation solar telescopes and space weather monitoring.

astro-ph.SR↗

arXiv2025

Continuous nonlinear adaptive experimental design with gradient flow

In computational inverse problems, the optimal experimental design (OED) problem seeks the best locations in time and space at which to take measurements. We investigate the nonlinear OED problem in the context of continuously-indexed design space for the measurements. In contrast to traditional approaches that select experiments from a finite measurement set, a continuous design space is often a better reflection of practical experimental options, where there is considerable flexibility concerning where and when to take measurements. The continuously-indexed space introduces computational challenges, and we address them by employing gradient-flow and optimal transport techniques, complemented by an adaptive strategy for bi-level optimization. Numerical results on the Lorenz 63 system and Schrodinger equation demonstrate that our solver identifies good measurement times / locations and achieves improved reconstruction of unknown parameters in inverse problems.

math.NA↗

arXiv2025

Monotonic Learning in the PAC Framework: A New Perspective

Monotone learning describes learning processes in which expected performance consistently improves as the amount of training data increases. However, recent studies challenge this conventional wisdom, revealing significant gaps in the understanding of generalization in machine learning. Addressing these gaps is crucial for advancing the theoretical foundations of the field. In this work, we utilize Probably Approximately Correct (PAC) learning theory to construct a theoretical risk distribution that approximates a learning algorithm's actual performance. We rigorously prove that this theoretical distribution exhibits monotonicity as sample sizes increase. We identify two scenarios under which deterministic algorithms based on Empirical Risk Minimization (ERM) are monotone: (1) the hypothesis space is finite, or (2) the hypothesis space has finite VC-dimension. Experiments on two classical learning problems validate our findings by demonstrating that the monotonicity of the algorithms' generalization error is guaranteed, as its theoretical risk upper bound monotonically converges to 0.

cs.LG↗

arXiv2025

Back-Projection Diffusion: Solving the Wideband Inverse Scattering Problem with Diffusion Models

We present Wideband Back-Projection Diffusion, an end-to-end probabilistic framework for approximating the posterior distribution induced by the inverse scattering map from wideband scattering data. This framework produces highly accurate reconstructions, leveraging conditional diffusion models to draw samples, and also honors the symmetries of the underlying physics of wave-propagation. The procedure is factored into two steps: the first step, inspired by the filtered back-propagation formula, transforms data into a physics-based latent representation, while the second step learns a conditional score function conditioned on this latent representation. These two steps individually obey their associated symmetries and are amenable to compression by imposing the rank structure found in the filtered back-projection formula. Empirically, our framework has both low sample and computational complexity, with its number of parameters scaling only sub-linearly with the target resolution, and has stable training dynamics. It provides sharp reconstructions effortlessly and is capable of recovering even sub-Nyquist features in the multiple-scattering regime.

cs.LG↗

arXiv2025

Correcting Noisy Multilabel Predictions: Modeling Label Noise through Latent Space Shifts

Noise in data appears to be inevitable in most real-world machine learning applications and would cause severe overfitting problems. Not only can data features contain noise, but labels are also prone to be noisy due to human input. In this paper, rather than noisy label learning in multiclass classifications, we instead focus on the less explored area of noisy label learning for multilabel classifications. Specifically, we investigate the post-correction of predictions generated from classifiers learned with noisy labels. The reasons are two-fold. Firstly, this approach can directly work with the trained models to save computational resources. Secondly, it could be applied on top of other noisy label correction techniques to achieve further improvements. To handle this problem, we appeal to deep generative approaches that are possible for uncertainty estimation. Our model posits that label noise arises from a stochastic shift in the latent variable, providing a more robust and beneficial means for noisy learning. We develop both unsupervised and semi-supervised learning methods for our model. The extensive empirical study presents solid evidence to that our approach is able to consistently improve the independent models and performs better than a number of existing methods across various noisy label settings. Moreover, a comprehensive empirical analysis of the proposed method is carried out to validate its robustness, including sensitivity analysis and an ablation study, among other elements.

cs.LG↗

arXiv2025

Constraints on violation of Lorentz symmetry with clock-comparison redshift experiments

Lorentz symmetry is a cornerstone of both the General relativity and Standard Model and its experimental verification deepens our understanding of nature. This paper focuses on the investigation of Lorentz violations with the context of clock comparison experiments in the framework of Standard Model Extension (SME). Considering matter-gravity coupling sector, we provide a generic frame to study the sensitivities of Lorentz-violating coefficients for three distinct types of clock redshift tests, including the traditional gravitational redshift test, null-redshift test I and null-redshift test II. Each of these tests is sensitivity to different combinations of Lorentz-violating coefficients. By using the current clock comparison results, we estimate the limits of SME coefficients at level of parts in $10^{4}$ down to parts in $10^{7}$. Better sensitivity may be achieved in the clock comparisons by using the state-of-the-art optical clocks. Additionally considering relativistic factors in null-redshift I, the frequency comparison result of E2 and E3 transitions of Yb$^{+}$ can set the limit $c^{e}_{00}=(7.4\pm9.3)\times10^{-9}$ in the electron sector. Our analysis demonstrates that clock-comparison redshift experiments may contribute to explore the vast parameters space on searching for the Lorentz violation.

gr-qc↗