Search arXiv⌕ Search

arXiv subjects

Gang Li

Publications and source records attributed to Gang Li.

At least 91 records · Page 5Linked to original sources
arXiv2026

Memory-Efficient Continual Learning with CLIP Models

Contrastive Language-Image Pretraining (CLIP) models excel at understanding image-text relationships but struggle with adapting to new data without forgetting prior knowledge. To address this, models are typically fine-tuned using both new task data and a memory buffer of past tasks. However, CLIP's contrastive loss suffers when the memory buffer is small, leading to performance degradation on previous tasks. We propose a memory-efficient, distributionally robust method that dynamically reweights losses per class during training. Our approach, tested on class incremental settings (CIFAR-100, ImageNet1K) and a domain incremental setting (DomainNet) adapts CLIP models quickly while minimizing catastrophic forgetting, even with minimal memory usage.

cs.LG↗
arXiv2026

Multi-fidelity surrogates for mechanics of composites: from co-kriging to multi-fidelity neural networks

Composite materials exhibit strongly hierarchical and anisotropic properties governed by coupled mechanisms spanning constituents, plies, laminates, structures, and manufacturing history. This intrinsic complexity makes predictive modeling of composites expensive, because repeated experiments and high-fidelity simulations are needed to cover large design spaces of material, structure, and manufacturing. Multi-fidelity surrogate modeling addresses this challenge by combining abundant, less expensive data with limited high-accuracy data to recover reliable high-fidelity predictions. This review presents a structured overview of multi-fidelity modeling for composite mechanics, covering Gaussian-process or Kriging-based methods, including co-Kriging, coregionalization models, autoregressive formulations, nonlinear autoregressive Gaussian processes, multi-fidelity deep Gaussian processes, and multi-fidelity neural networks. Their distinctions are examined in terms of cross-fidelity correlation, discrepancy representation, uncertainty quantification, and scalability. Selected examples of their applications to composites are introduced according to the roles that multi-fidelity surrogates play in engineering problems, including forward prediction for rapid exploration of material design spaces, inverse optimization for composite parameter identification and design search under limited high-fidelity access, and workflow integration, where heterogeneous data sources, constraints, and validation requirements determine model utility. Open question discussions highlight recurring challenges specific to composites, such as regime-dependent fidelity gaps associated with nonlinear damage and manufacturing history, mismatches between simulations and experiments, and uncertainty propagation across multi-fidelity models.

physics.comp-ph↗
arXiv2026

Two distinct superconducting regimes in Ti4Co2O under pressures

We report on the pressure dependence of superconducting transition temperature Tc and upper critical field Bc2(0) through electrical transport of the Ti4Co2O superconductor (eg.,the superconducting transition temperature Tc = 2.5 K and the Bc2(0)=7.2T=2.9Tc). We find that the Tc exhibits non-monotonic pressure dependence:it rises monotonically at first with a pressure coefficient of dTc/dP=0.034 K/GPa, but rapidly decreases around 10-20 GPa, and then increases with the dTc/dP = 0.023 K/GPa, up to= 4.31 K at 69.7 GPa. Concurrently, the Bc2(0)exhibits a dome shaped pressure dependence, with its maximum at 5 GPa of almost twice the value at ambient pressure, exceeding the weak-coupling Pauli paramagnetic limit Bp throughout the whole pressure range. By comparing the normal-state and superconducting properties, we identify two distinct superconducting regimes, with a low-pressure superconducting phase characterized by an enhanced Bc2(0)values and Fermi-liquid normal-state electrical transport (the exponent n = 2), and a high-pressure superconducting phase with a monotonically increased Tc and an enhancement in phonon scatterings (the exponent n = 4). Room-temperature synchrotron X-ray diffraction indicates that there is no structural transition up to 55.8 GPa, which gives a relatively large bulk modulus of 192 GPa in comparison with other alloy superconductors. First-principles calculations suggest that the nonmonotonic Tc maybe closely related to the evolution of the density of states of Ti4Co2O upon compression, which is different from those of isostructural superconductors Ti4Ir2O and Nb4Rh2C. Our results show that even in the Ti4Co2O with weak spin-orbit coupling, superconductivity remains highly sensitive to the external stimuli such as pressure.

cond-mat.supr-con↗
arXiv2026

Hunting for $B\bar B$ molecular state $X_{b0}$ via radiative transition of $Υ(10753)$

We investigate the radiative decay $Υ(10753) \to γX_{b0}$ within the framework of nonrelativistic effective field theory (NREFT). The $Υ(10753)$ is treated as an $S$-$D$ mixed state of the $Υ(4S)$ and $\YD$, while the $X_{b0}$ is interpreted as a weakly bound $B\bar{B}$ molecule with $J^{PC}=0^{++}$. The decay process was assumed to occur via the intermediate meson loops involving the $S$-wave $B^{(*)}$ and $P$-wave $B_1^{(\prime)}$ mesons. Our calculated results indicate that the decay $Υ(10753) \to γX_{b0}$ is dominated by the $B_1^{(\prime)}$ meson loops. The partial decay width is predicted to be $0.2-1.5~\mathrm{keV}$ for a binding energy range of $ε_X=0-10~\mathrm{MeV}$, corresponding to a branching fraction of $10^{-6}-10^{-5}$. The decay width is found to be insensitive to the full width of the $B_1^{\prime}$ meson. Our study suggests that the radiative decay of the $Υ(10753)$ is a promising channel to search for the $X_{b0}$ state, which is crucial for testing heavy-quark symmetries and understanding the exotic hadron spectrum in the bottom sector.

hep-ph↗
arXiv2026

Probabilistic Predictions of Process-Induced Deformation in Carbon/Epoxy Composites Using a Deep Operator Network

Fiber reinforcement and polymer matrix respond differently to manufacturing conditions due to mismatch in coefficient of thermal expansion and matrix shrinkage during curing of thermosets. These heterogeneities generate residual stresses over multiple length scales, whose partial release leads to process-induced deformation (PID), requiring accurate prediction and mitigation via optimized non-isothermal cure cycles. This study considers a unidirectional AS4 carbon fiber/amine bi-functional epoxy prepreg and models PID using a two-mechanism framework that accounts for thermal expansion/shrinkage and cure shrinkage. The model is validated against manufacturing trials to identify initial and boundary conditions, then used to generate PID responses for a diverse set of non-isothermal cure cycles (time-temperature profiles). Building on this physics-based foundation, we develop a data-driven surrogate based on Deep Operator Networks (DeepONets). A DeepONet is trained on a dataset combining high-fidelity simulations with targeted experimental measurements of PID. We extend this to a Feature-wise Linear Modulation (FiLM) DeepONet, where branch-network features are modulated by external parameters, including the initial degree of cure, enabling prediction of time histories of degree of cure, viscosity, and deformation. Because experimental data are available only at limited time instances (for example, final deformation), we use transfer learning: simulation-trained trunk and branch networks are fixed and only the final layer is updated using measured final deformation. Finally, we augment the framework with Ensemble Kalman Inversion (EKI) to quantify uncertainty under experimental conditions and to support optimization of cure schedules for reduced PID in composites.

cs.CE↗
arXiv2026

Certain Head, Uncertain Tail: Expert-Sample for Test-Time Scaling in Fine-Grained MoE

Test-time scaling improves LLM performance by generating multiple candidate solutions, yet token-level sampling requires temperature tuning that trades off diversity against stability. Fine-grained MoE, featuring hundreds of well-trained experts per layer and multi-expert activation per token, offers an unexplored alternative through its rich routing space. We empirically characterize fine-grained MoE routing and uncover an informative pattern: router scores exhibit a certain head of high-confidence experts followed by an uncertain tail of low-confidence candidates. While single-run greedy accuracy remains stable when fewer experts are activated, multi-sample pass@n degrades significantly-suggesting that the certain head governs core reasoning capability while the uncertain tail correlates with reasoning diversity. Motivated by these findings, we propose Expert-Sample, a training-free method that preserves high-confidence selections while injecting controlled stochasticity into the uncertain tail, enabling diverse generation without destabilizing outputs. Evaluated on multiple fine-grained MoE models across math, knowledge reasoning, and code tasks, Expert-Sample consistently improves pass@n and verification-based accuracy. On Qwen3-30B-A3B-Instruct evaluated on GPQA-Diamond with 32 parallel samples, pass@32 rises from 85.4% to 91.9%, and accuracy improves from 59.1% to 62.6% with Best-of-N verification.

cs.LG↗
arXiv2026

Electrically-controllable superconducting memory effect in UTe2

If a computer could be assembled from superconducting components, the energy efficiency would far surpass that of conventional electronics. Historic research efforts towards this goal yielded pivotal breakthroughs in the development and discovery of scanning tunnelling microscopy and high temperature superconductivity. Although recent strides have been taken in advancing superconducting diode and switching technologies, harnessing read/writeable memory functionality in superconducting platforms has remained challenging. Here we show that bulk single crystal specimens of the triplet superconductor candidate uranium ditelluride (UTe$_2$) possess such properties. Upon applying a magnetic field to access an intermediate regime straddling two distinct superconducting phases, we find that direct current pulses can push the material in and out of a metastable state possessing an enhanced critical current $J_c$. This switching is controllable by the strength and duration of the stimuli, with the system `remembering' whether it is in the high or low $J_c$ state for extended periods. We interpret this to be due to competition between two distinct vortex species, which can be perturbatively pushed into a non-equilibrium high-disorder configuration with stronger pinning forces and thus higher $J_c$. Rather than requiring proximate magnetic or semiconducting interfaces, this memory functionality appears to be an intrinsic property of UTe$_2$ rooted in the superconducting order itself. Our findings underscore the rich complexity of quantum vortex matter, and demonstrate the viability of engineering a new class of superconducting memory elements with ultralow-power switching.

cond-mat.supr-con↗
arXiv2026

A Loewner-Nirenberg phenomena for Ricci flow on compact manifolds with boundary

In this paper, we show that starting from a geodesic ball $\overline{B_{r_0}}(0)$ in $\mathbb{H}^n$, for $n\geq3$, with prescribed non-decreasing rotationally symmetric mean curvature and the fixed conformal class $[g_{\mathbb{S}^{n-1}}]$ on the boundary, the solution $g(t)$ to the normalized Ricci flow $(1.2)$ which is continuous up to the boundary, exists for all $t>0$ and converges locally uniformly in $B_{r_0}(0)$ to a complete hyperbolic metric as $t\to\infty$(see Theorem 1.2 for details). Moreover, the sectional curvature of $g(t)$ maintains less than $-1$ for $t>0$. For dimension $2$, to achieve such a convergence result, we need the additional assumption that the mean curvature on the boundary increases in a certain speed to infinity as $t\to\infty$.

math.DG↗
arXiv2026

Probing Red Giant Interiors with G-Dominated Mixed Modes I: The Cases of KIC 9145955, KIC 9970396, KIC 9882316 and KIC 11968334

We perform a detailed asteroseismic analysis of four red giants observed by Kepler: KIC 9145955, KIC 9970396, KIC 9882316, and KIC 11968334. Our study is based on individual oscillation frequencies, with particular emphasis on gravity-dominated (g-dominated) mixed modes. These modes are highly sensitive to the deep stellar interior and serve as powerful diagnostics of core structure, convective overshooting, and internal rotation. Moreover, surface effects have minimal impact on g-dominated mixed modes. To ensure accurate frequency matching between observations and theoretical models, we apply a mode-identification technique that effectively distinguishes p-dominated from g-dominated modes. Although a definitive confirmation of this trend requires a substantially larger asteroseismic sample, our best-fitting models suggest that the derived convective overshooting parameter ($f_{ov}$) increases with stellar mass. In particular, within our sample the star with a mass exceeding $1.4M_{\odot}$ requires $f_{ov} > 0.01$, whereas lower-mass red giants tend to have $f_{ov}$ <0.01. In addition, the average core rotation rate of KIC 11968334 is precisely determined to be $0.7409\pm0.0113 μ$Hz from the asteroseismic model.

astro-ph.SR↗
arXiv2026

Sub-spin-flop switching of a fully compensated antiferromagnet by magnetic field

The control of antiferromagnets by magnetic fields represents a fundamental challenge in condensed matter physics, owing to their fully compensated magnetic order and vanishing net magnetization. Conventional methods rely on either uncompensated moments or high-field spin-flop transitions. Here, we demonstrate low-field switching in the fully compensated antiferromagnet CeNiAsO -- a material recently proposed as a candidate for $p$-wave magnetism. Using an in-plane magnetic field well below the spin-flop threshold, we selectively stabilize one of two degenerate antiferromagnetic domains with mutually orthogonal sublattice orientations. This field-induced domain selection allows reversible and nonvolatile switching of a giant in-plane resistivity anisotropy up to $\sim35\,\%$ -- a magnitude that far exceeds conventional anisotropy signals driven by spin-orbit coupling. The switching behavior persists across both the low-temperature noncollinear Néel phase and the higher-temperature collinear spin-density-wave phase, highlighting the universality of the domain-selection mechanism. Our work establishes a practical approach for manipulating compensated antiferromagnets with modest magnetic fields and underscores their potential for high-performance spintronic devices based on giant and switchable resistivity anisotropy.

cond-mat.mes-hall↗
arXiv2026

MedVerse: Efficient and Reliable Medical Reasoning via DAG-Structured Parallel Execution

Large language models (LLMs) have demonstrated strong performance and rapid progress in a wide range of medical reasoning tasks. However, their sequential autoregressive decoding forces inherently parallel clinical reasoning, such as differential diagnosis, into a single linear reasoning path, limiting both efficiency and reliability for complex medical problems. To address this, we propose MedVerse, a reasoning framework for complex medical inference that reformulates medical reasoning as a parallelizable directed acyclic graph (DAG) process based on Petri net theory. The framework adopts a full-stack design across data, model architecture, and system execution. For data creation, we introduce the MedVerse Curator, an automated pipeline that synthesizes knowledge-grounded medical reasoning paths and transforms them into Petri net-structured representations. At the architectural level, we propose a topology-aware attention mechanism with adaptive position indices that supports parallel reasoning while preserving logical consistency. Systematically, we develop a customized inference engine that supports parallel execution without additional overhead. Empirical evaluations show that MedVerse improves strong general-purpose LLMs by up to 8.9%. Compared to specialized medical LLMs, MedVerse achieves comparable performance while delivering a 1.3x reduction in inference latency and a 1.7x increase in generation throughput, enabled by its parallel decoding capability. Code is available at https://github.com/aiming-lab/MedVerse.

cs.LG↗
arXiv2026

Variability classification of TESS targets in LOPS2, the first long-term pointing field of PLATO. Version 1 of the public variability catalogue

The PLAnetary Transits and Oscillations of stars (PLATO) mission is expected to launch in January 2027. A total of 8\% of its data rate will be dedicated to complementary science targets selected from approved Guest Observer proposals. We seek to provide an open-source catalogue of variable stars in PLATO's first long-term observing field, LOPS2. We want to use existing observations from the Transiting Exoplanet Survey Satellite (TESS), which has observed many stars in LOPS2. We classified 38 million calibrated aperture light curves from the TESS-Gaia Light Curve pipeline (TGLC, $G\lesssim17$) for 6 million unique sources in LOPS2 with two machine learning frameworks -- a deep neural network and a feature-based gradient-boosted decision-tree ensemble. We combined their predictions to create this first version of the LOPS2 variability catalogue, performed manual vetting of a sub-sample classified light curves, and a statistical analysis of the results to validate our methodology and to assess the variability properties and parameters of the stars in the catalogue. Our classification resulted in the identification of approximately 72% of the light curves having dominant instrument- or pipeline-induced signal, with the remaining 28% representing 3.6 million individual candidate variable stars, including pulsating, rotating, and eclipsing stars. Candidate pulsators exhibit varied behaviour in terms of their frequencies, amplitudes, rotation, and fundamental parameters. To ensure purity of the samples, filtering on colour, luminosity, the dominant frequency and its amplitude, and presence of close neighbours is helpful. We provide the first version of our PLATO LOPS2 variability catalogue to the community for further study and scrutiny. It is to date one of the largest catalogues of variable stars from an automated classification pipeline.

astro-ph.SR↗
arXiv2026

Study of $χ_{cJ}\to ηηη^\prime$ via intermediate charmed meson loop mechanisms and its implications for non-observation of $η_1(1855)$ in $χ_{cJ}$ decays

Recently, the BESIII Collaboration reported the first observation of the decays $χ_{cJ} \to ηηη^\prime$ in order to search for the $1^{-+}$ exotic state $η_1(1855)$. A partial wave analysis of the $ηη^\prime$ invariant mass spectrum shows no significant signal for the $η_1(1855)$. In this work, we, using an effective Lagrangian approach, investigate the processes $χ_{cJ} \to ηηη^\prime$ via the box and triangle loops involving charmed mesons and the scalar meson $f_0(1500)$. Our calculations reproduce well the experimental branching fractions of $χ_{cJ} \to ηηη^\prime$. Furthermore, we present the predictions of the relevant invariant mass spectra of $ηη^\prime$ and $ηη$ produced in the $χ_{c1}$ decay, which seem overall consistent with the BESIII measurements. In the present model, the decay $χ_{c1} \to ηηη^\prime$ is dominated by the triangle and box loop contributions. The consistency between our theoretical results and the BESIII measurements sheds light on the underlying decay mechanism of the $χ_{cJ}$ decaying into light mesons and might be helpful to understand the absence of the $η_1(1855)$ signal in the decay channels $χ_{cJ} \to ηηη^\prime$.

hep-ph↗
arXiv2026

On A Class of Greedy Sparse Recovery Algorithms

Sparse signal recovery deals with finding the sparsest solution of an under-determined linear system $\vx = \mQ\vs$. In this paper, we propose a novel greedy approach to addressing the challenges from such a problem. Such an approach is based on a characterization of solutions to the system, which allows us to work on the sparse recovery in the $\vs$-space directly with a given measure. With $l_2$-based measure, an orthogonal matching pursuit (OMP)-type algorithm is proposed, which significantly outperforms the classical OMP algorithm in terms of recovery accuracy while maintaining comparable computational complexity. An $l_1$-based algorithm, denoted as $\text{Alg}_{GL1}$, is derived. Such an algorithm significantly outperforms the classical basis pursuit (BP) algorithm. Combining with the CoSaMP-strategy for selecting atoms, a class of high performance greedy algorithms is also derived. Extensive numerical simulations on both synthetic and image data are carried out, with which the superior performance of our proposed algorithms is demonstrated in terms of sparse recovery accuracy and robustness against numerical instability of the system matrix $\mQ$ and disturbance in the measurement $\vx$.

cs.IT↗
arXiv2026

Isochrone-cloud fitting and asteroseismology of the Kepler open cluster NGC6866

We investigate how isochrones computed with different input physics and initial conditions affect the age dating of the open cluster NGC 6866, and compare the results with asteroseismic ages derived from Kepler photometry. Using Gaia DR3 data, we identified 180 cluster members with a clustering algorithm. We then developed an isochrone-cloud fitting method that accounts for a range of free parameters in the input physics. Variable stars were subsequently identified among the cluster members. For 19 g-mode pulsators, we carried out modelling with a dedicated grid of rotating stellar models, constrained by spectroscopic and photometric parameters, the asymptotic gravity-mode period spacing, and the near-core rotation rate. We considered two cases: modelling each pulsator individually and modelling them under the assumption of a common cluster age. PARSEC and MIST isochrones yield discrepant ages of 690 and 467 Myr, respectively. The isochrone-cloud fit indicates an initial critical rotation distribution peaking at 0.6, about a factor of two higher than inferred from asteroseismology. The seismic modelling shows agreement between seismic and isochronal masses, but substantial differences in the derived ages due to differences in internal mixing. When the g-mode pulsators are modelled with a shared cluster age, we obtain 759 Myr, consistent with the PARSEC isochronal age. We conclude that age dating of open clusters is sensitive to the adopted input physics and initial conditions, highlighting the need for better calibrated stellar evolutionary models.

astro-ph.SR↗
arXiv2026

Route-Induced Density and Stability (RIDE): Controlled Intervention and Mechanism Analysis of Routing-Style Meta Prompts on LLM Internal States

Routing is widely used to scale large language models, from Mixture-of-Experts gating to multi-model/tool selection. A common belief is that routing to a task ``expert'' activates sparser internal computation and thus yields more certain and stable outputs (the Sparsity--Certainty Hypothesis). We test this belief by injecting routing-style meta prompts as a textual proxy for routing signals in front of frozen instruction-tuned LLMs. We quantify (C1) internal density via activation sparsity, (C2) domain-keyword attention, and (C3) output stability via predictive entropy and semantic variation. On a RouterEval subset with three instruction-tuned models (Qwen3-8B, Llama-3.1-8B-Instruct, and Mistral-7B-Instruct-v0.2), meta prompts consistently densify early/middle-layer representations rather than increasing sparsity; natural-language expert instructions are often stronger than structured tags. Attention responses are heterogeneous: Qwen/Llama reduce keyword attention, while Mistral reinforces it. Finally, the densification--stability link is weak and appears only in Qwen, with near-zero correlations in Llama and Mistral. We present RIDE as a diagnostic probe for calibrating routing design and uncertainty estimation.

cs.AI↗
arXiv2026

Role of electromagnetic corrections in the $ππ$ distributions of $ψ^\prime \to J/ψππ$

The cusp structure at the $π^+π^-$ threshold in the $π^0π^0$ invariant mass spectrum serves as a sensitive probe for extracting the $S$-wave $ππ$ scattering lengths in processes where an $S$-wave $π^0π^0$ pair is produced in the final states. Within the framework of nonrelativistic effective field theory with coupled channels $π^0π^0$ and $π^+π^-$, we revisit the near-threshold structures in the $π^0π^0$ spectrum of $ψ^\prime \to J/ψππ$. Our analysis incorporates the $ππ$ final-state rescattering, including both strong and Coulomb interactions. It turns out that the cusp near the $π^+π^-$ threshold becomes more prominent when Coulomb interactions are included. The electromagnetic correctionsare found to alter the magnitude of the threshold cusp by about 2%-3%, underscoring the necessity of including these effects in precision determinations of the $ππ$ scattering lengths. The coupled-channel amplitude constructed in this work provides a ready-to-use theoretical framework for experimental analyses of fine structures near $ππ$ thresholds.

hep-ph↗
arXiv2026

MiDashengLM: Efficient Audio Understanding with General Audio Captions

Current approaches for large audio language models (LALMs) often rely on closed data sources or proprietary models, limiting their generalization and accessibility. This paper introduces MiDashengLM, a novel open audio-language model designed for efficient and comprehensive audio understanding through the use of general audio captions using our novel ACAVCaps training dataset. MiDashengLM exclusively relies on publicly available pretraining and supervised fine-tuning (SFT) datasets, ensuring full transparency and reproducibility. At its core, MiDashengLM integrates Dasheng, an open-source audio encoder, specifically engineered to process diverse auditory information effectively. Unlike previous works primarily focused on Automatic Speech Recognition (ASR) based audio-text alignment, our strategy centers on general audio captions, fusing speech, sound and music information into one textual representation, enabling a holistic textual representation of complex audio scenes. Lastly, MiDashengLM provides an up to 4x speedup in terms of time-to-first-token (TTFT) and up to 20x higher throughput than comparable models. Checkpoints are available online at https://huggingface.co/mispeech/midashenglm-7b and https://github.com/xiaomi-research/dasheng-lm.

cs.SD↗