Searcharxiv⌕ Search

arXiv subjects

Ran Wang

Publications and source records attributed to Ran Wang.

At least 37 records · Page 2Linked to original sources

Proton-electron coupled catalyst for ionomer-free electrochemical energy conversion

Efficient electrochemical energy devices are vital to renewable energy technology, yet coordinating the effective flow of electrons, ions, and chemical species continues to be a major challenge. In conventional proton-exchange membrane fuel cell (PEMFC) catalyst layers, proton and electron transport are supplied separately through percolating carbon networks and ionomer binders, rendering the catalyst largely passive and imposing fundamental trade-offs between reactant accessibility, ionic conductivity, and catalyst activity. Here, we introduce a one-dimensional proton-electron coupled catalyst (PECC) design, a transport-integrated electrocatalyst architecture in which the catalyst itself simultaneously supplies electronic and protonic transport to catalyst active sites. Using this PECC, PEMFCs can have an ionomer-free cathode catalyst layer (CCL), resulting in a dramatic 95% reduction in non-Fickian oxygen transport and boosting power density by 34% and 85% compared to traditional CCLs, with cathode Pt loadings of approximately 0.090 mg/cm^2 and 0.037 mg/cm^2, respectively. Meanwhile, PECC retains 65% of its mass activity and exhibits 32% higher power density than its ionomer-based CCL counterpart after 30k accelerated stressed test. Similar mass transport improvements have been observed in the electrochemical hydrogen pump (EHP) using PECC in the catalyst layers. Molecular dynamics simulations show the PECC's proton conductivity is 249% higher than Nafion. This PECC catalyst structure addresses core transport problems in PEMFCs, leading to almost 20% improvement in fuel efficiency and opens up new possibilities for designing high-performance, cost-effective electrochemical devices.

cond-mat.mtrl-sci↗

Strassen's local law of the iterated logarithm for the generalized fractional Brownian motion

Let $X:=\{X(t)\}_{t\ge0}$ be a generalized fractional Brownian motion given by $$ \{X(t)\}_{t\ge0}\overset{d}{=}\left\{ \int_{\mathbb R} \left((t-u)_+^α-(-u)_+^α \right) |u|^{-γ/2} B(du) \right\}_{t\ge0}, $$ with parameters $γ\in (0, 1)$ and $α\in \left(-1/2+ γ/2, \, 1/2+γ/2\right)$. This process was introduced by Pang and Taqqu (2019) as the scaling limit of a class of power-law shot noise processes. The parameters $α$ and $γ$ govern the probabilistic and statistical properties of $X$. In particular, the parameter $γ$ breaks the stationarity of increments of $X$. In this paper, we establish Strassen's local law of the iterated logarithm for $X$ at a given point $t_0 \in (0, \infty)$. This result describes explicitly the roles played by the parameters $α, γ$, and the location $t_0$. Our theorem differs from the earlier Strassen's {global law of the iterated logarithm} for $X$ proved by Ichiba, Pang and Taqqu (2022).

math.PR↗

CO(7-6) and [C I](2-1) survey in z > 6 quasars

High-redshift ($z\gtrsim6$) quasars trace the earliest supermassive black holes and intense star formation, offering key laboratories for black hole-galaxy evolution at cosmic dawn. While far-infrared studies have revealed large dust reservoirs and strong [C II] emission, the physical conditions and molecular gas content of their ISM remain uncertain. We present ALMA Band 3 observations of the redshifted CO(7-6) and [C I](2-1) emission lines and dust continuum in a sample of 18 quasars at $z \sim 6$. We detected CO(7-6) in 15/18, [C I](2-1) in 6/18, and continuum in 13/18 sources. Line luminosities and continuum fluxes were used to estimate molecular gas masses from CO, [C I], and dust, and a hierarchical Bayesian cross-calibration of all four tracers yielded consistent per-source $M_{\rm H_2}$ estimates and conversion factors. Comparison with PDR and XDR model grids using the $L_{\rm [CII]}/L_{\rm [CI]}$ and $L_{\rm CO(7--6)}/L_{\rm TIR}$ ratios suggests gas densities of $n > 10^4$ cm$^{-3}$ and radiation fields of $G_0 \sim 10^3$--$10^4$ for sources consistent with PDR solutions, while many quasars fall outside the model parameter space. The $L'_{\rm CO(7-6)}/L'_{\rm [CI](2-1)}$ ratio indicates that a large fraction of the molecular gas resides in a warm and highly excited phase. Together these results suggest that classical PDR heating alone cannot explain the observed line ratios and that additional volumetric processes such as X-ray irradiation, turbulence and shocks, or enhanced cosmic-ray heating likely influence the excitation of the cold ISM. They demonstrate the power of multi-line diagnostics in revealing the excitation and structure of the cold ISM in early quasar host galaxies and highlight the need for joint analysis of CO, [C I], [C II], and dust emission to characterize star formation and AGN-driven heating at cosmic dawn.

astro-ph.GA↗

Finding the Correct Visual Evidence Without Forgetting: Mitigating Hallucination in LVLMs via Inter-Layer Visual Attention Discrepancy

Large Vision-Language Models (LVLMs) have shown remarkable performance on a wide range of vision-language tasks. Despite this progress, they are still prone to hallucination, generating responses that are inconsistent with visual content. In this work, we find that LVLMs tend to hallucinate when they pay insufficient attention to the correct visual evidence and gradually forget it during the generation process. We empirically find that although LVLMs overall attend insufficiently to visual evidence, they exhibit sensitivity to the correct visual evidence in specific layers, with notable inter-layer discrepancy. Motivated by this observation, we propose a novel hallucination mitigation method that enhances visual evidence based on Inter-Layer Visual Attention Discrepancy (ILVAD). Specifically, we obtain the attention weights from early generated tokens to visual tokens across layers and identify the tokens that are repeatedly activated as visual evidence, forming a saliency map. We then enhance attention to visual evidence during generation through the saliency map to reduce visual forgetting. In addition, we leverage the saliency map to obtain attention scores of generated text to visual evidence, in order to select and emphasize text tokens that are strongly grounded in visual evidence. Our method is training-free and plug-and-play. Multiple benchmark evaluations conducted on five recently released models show that our method can consistently mitigate hallucinations in different LVLMs over various architectures. Code is available at https://github.com/ytx-ML/ILVAD.

cs.CV↗

Growth rates for the Hölder coefficients of the linear stochastic fractional heat equation with rough dependence in space

We study the linear stochastic fractional heat equation $$ \frac{\partial}{\partial t}u(t,x)=-(-Δ)^{\fracα2}u (t,x)+\dot{W}(t,x),\ \ t> 0,\ \ x\in\RR, $$ where $-(-Δ)^{\fracα{2}}$ denotes the fractional Laplacian with power $α\in (1, 2)$, and the driving noise $\dot W$ is a centered Gaussian field which is white in time and has the covariance of a fractional Brownian motion with Hurst parameter $H\in\left(\frac {2-α}2,\frac 12\right)$. We establish exact asymptotics for the solution as both time and space variables tend to infinity and derive sharp growth rates for the Hölder coefficients. The proofs are based on Talagrand's majorizing measure theorem and Sudakov's minoration theorem.

math.PR↗

CAX-Agent: A Lightweight Agent Harness for Reliable APDL Automation

Large language models deployed for MAPDL finite-element simulation face practical reliability challenges: without structured execution control, tool encapsulation, and fault recovery, outputs may be inconsistent and task failures are common. The Agent Harness paradigm addresses this by inserting domain-specific orchestration middleware that manages tool lifecycles, workflow state, and recovery escalation. This paper presents the architecture of CAX-Agent, a lightweight agent harness purpose-built for MAPDL automation, and empirically evaluates one of its core components -- the recovery policy.CAX-Agent organizes execution into three layers -- LLM service, agent harness, and solver backend -- with a recovery ladder that escalates from deterministic rule patching through model-driven regeneration to context enrichment and human intervention. We evaluate three recovery strategies (no_recovery, rule_only, and model_only) on 50 standard structural benchmarks with three repeated runs per strategy (450 case-runs total). Two independent human raters score task completion under blind conditions; inter-rater agreement is strong (quadratic weighted Cohen's kappa = 0.84, 96 percent of score pairs within one point). Model_only achieves the best completion rate (0.9267), task score (3.59/4), total score (9.16/10), and zero-intervention rate (0.84), outperforming rule_only (0.7733, 3.17/4, 7.03/10, 0.00) and no_recovery (0.6933, 2.74/4, 5.60/10, 0.00) with large effect sizes (Cliff's delta = 0.81-0.87). The benchmark uses deliberately simple geometries to isolate recovery-policy effects; we discuss the scope of these findings and directions for broader validation.

cs.AI↗

Moderate Deviation Principle for dynamical systems with small random perturbation

Consider the stochastic differential equation in $\rr^d$ dX^{\e}_t&=b(X^{\e}_t)dt+\sqrt{\e}σ(X^\e_t)dB_t X^{\e}_0&=x_0,\quad x_0\in\rr^d$ where $b:\rr^d\to\rr^d$ is $C^1$ such that $ \leq C(1+|x|^2)$, $σ:\rr^d\to \MM(d\times n)$ is locally Lipschitzian with linear growth, and $B_t$ is a standard Brownian motion taking values in $\rr^n$. Freidlin-Wentzell's theorem gives the large deviation principle for $X^\e$ for small $\e$. In this paper we establish its moderate deviation principle.

math.PR↗

Stochastic fractional heat equation with general rough noise

Consider the following nonlinear one-dimensional stochastic fractional heat equation $$\frac{\partial }{\partial t}u(t, x)= -(-Δ)^{α/2}u(t, x) +σ(t,x,u(t,x)) \dot{W}(t, x), $$ where $-(-Δ)^{α/2}$ is the fractional Laplacian on $\mathbb R$ for $1<α<2$, and $\dot{W}$ is a Gaussian noise that is white in time and behaves in space as a fractional Brownian motion with Hurst index $H$ satisfying $\frac{3-α}{4}<H<\frac12$. When $α=2$, Hu and Wang ({\it Ann. Inst. Henri Poincaré Probab. Stat.} {\bf 58} (2022) 379-423) studied the well-posedness of the solution and its Hölder continuity, removing the technical condition $σ(0)=0$ that was previously assumed in Hu et al. ({\it Ann. Probab.} {\bf 45} (2017) 4561-4616). Their approach relied on working in a weighted space with a suitable power decay function. For the case $α\in (1,2)$, inspired by Hu and Wang, we investigate the well-posedness of the stochastic fractional heat equation without imposing the technical condition of $σ(0)=0$, which was required in the earlier work of Liu and Mao ({\it Bull. Sci. Math.} {\bf181} (2022) 103207). In our analysis, precise estimates of the heat kernel associated with the fractional Laplacian $-(-Δ)^{α/2}$ play a crucial role.

math.PR↗

PandaPose: 3D Human Pose Lifting from a Single Image via Propagating 2D Pose Prior to 3D Anchor Space

3D human pose lifting from a single RGB image is a challenging task in 3D vision. Existing methods typically establish a direct joint-to-joint mapping from 2D to 3D poses based on 2D features. This formulation suffers from two fundamental limitations: inevitable error propagation from input predicted 2D pose to 3D predictions and inherent difficulties in handling self-occlusion cases. In this paper, we propose PandaPose, a 3D human pose lifting approach via propagating 2D pose prior to 3D anchor space as the unified intermediate representation. Specifically, our 3D anchor space comprises: (1) Joint-wise 3D anchors in the canonical coordinate system, providing accurate and robust priors to mitigate 2D pose estimation inaccuracies. (2) Depth-aware joint-wise feature lifting that hierarchically integrates depth information to resolve self-occlusion ambiguities. (3) The anchor-feature interaction decoder that incorporates 3D anchors with lifted features to generate unified anchor queries encapsulating joint-wise 3D anchor set, visual cues and geometric depth information. The anchor queries are further employed to facilitate anchor-to-joint ensemble prediction. Experiments on three well-established benchmarks (i.e., Human3.6M, MPI-INF-3DHP and 3DPW) demonstrate the superiority of our proposition. The substantial reduction in error by $14.7\%$ compared to SOTA methods on the challenging conditions of Human3.6M and qualitative comparisons further showcase the effectiveness and robustness of our approach.

cs.CV↗

Spin-Orbital Intertwined Topological Superconductivity in a Class of Correlated Noncentrosymmetric Materials

In this study, we propose an alternative route to achieving topological superconductivity (TSC). Our approach applies to a new class of correlated noncentrosymmetric materials that host two spin-split Fermi surfaces with identical spin textures due to a spin-orbital intertwined effect. Incorporating multi-orbital repulsive Hubbard interactions, we calculate the superconducting pairings of a minimal two-orbital effective model within a spin-fluctuation-mediated superconductivity framework. We find that, depending on the effective Rashba spin-orbit coupling (RSOC) strength and filling level, the Hubbard interaction can drive the leading pairing symmetry into the $A_1(S_{\pm})$, $B_1$, $B_2$ or $B_2(d_{\pm})$ irreducible representations (IRs) of the $C_{4v}$ point group. Notably, the $A_1(S_{\pm})$ pairing gives rise to a fully gapped TSC characterized by a $Z_2$ invariant, while the $B_2(d_{\pm})$ pairing results in a nodal TSC. Our analysis reveals that the fully gapped TSC is predominated by spin-singlet regardless of the presence of the spin-triplet components. This distinguishes our model from noncentrosymmetric materials with conventional Rashba-split band structures, where TSC typically emerges near the van Hove singularity and is primarily driven by $p$-wave or $f$-wave spin-triplet pairing. These features enhances its experimental accessibility, and we discuss potential experimental systems for its realization.

cond-mat.supr-con↗

Evidence of Gas Depletion in Quasars with Moderate Radio Emission

The energy released by active galactic nuclei (AGNs) is considered to have a profound impact on the cold gas properties of their host galaxies, potentially heating or removing the gas and further suppressing star formation. To understand the feedback from AGN radio activity, we investigate its impacts on the cold gas reservoirs in AGNs with different radio activity levels. We construct a quasar sample with a mean $z\sim1.5$ and a mean $L_{\rm bol}\sim10^{45.8}\ \rm erg\ s^{-1}$, all with Herschel detections to enable estimates of the total gas mass through the galactic dust continuum emission. The sample is then cross-matched with radio catalogs and divided into radio loud (RL) quasars, radio-detected radio quiet (RQ) quasars and radio-undetected quasars based on their radio loudness. Through spectral energy distribution (SED) fitting, we find the radio-detected RQ quasars exhibit evidence of gas deficiency with host galaxies possessing $\sim 0.3$ dex lower dust and gas masses compared to the other two groups, despite being matched in $M_{\rm BH}$, $L_{\rm bol}$, $M_{*}$ and SFR. Furthermore, evidence from optical spectra shows that both the fraction and velocity of outflows are higher in the radio-detected RQ group, suggesting a connection between the ionized gas outflows and the moderate radio activity. These results suggest that the AGN feedback could be more efficient in AGNs with weak/moderate radio emission than in those without radio detection or those with strong radio emission. Further high-resolution observations are needed to understand the interaction between the interstellar medium and the weak/moderate AGN radio activity.

astro-ph.GA↗

Numina-Lean-Agent: An Open and General Agentic Reasoning System for Formal Mathematics

Agentic systems have recently become the dominant paradigm for formal theorem proving, achieving strong performance by coordinating multiple models and tools. However, existing approaches often rely on task-specific pipelines and trained formal provers, limiting their flexibility and reproducibility. In this paper, we propose the paradigm that directly uses a general coding agent as a formal math reasoner. This paradigm is motivated by (1) A general coding agent provides a natural interface for diverse reasoning tasks beyond proving, (2) Performance can be improved by simply replacing the underlying base model, without training, and (3) MCP enables flexible extension and autonomous calling of specialized tools, avoiding complex design. Based on this paradigm, we introduce Numina-Lean-Agent, which combines Claude Code with Numina-Lean-MCP to enable autonomous interaction with Lean, retrieval of relevant theorems, informal proving and auxiliary reasoning tools. Using Claude Opus 4.5 as the base model, Numina-Lean-Agent solves all problems in Putnam 2025 (12 / 12), matching the best closed-source system. Beyond benchmark evaluation, we further demonstrate its generality by interacting with mathematicians to successfully formalize the Brascamp-Lieb theorem. We release Numina-Lean-Agent and all solutions at https://github.com/project-numina/numina-lean-agent.

cs.AI↗

Local linearization for the nonlinear damped stochastic Klein-Gordon equation

For the $1+1$ dimensional nonlinear damped stochastic Klein-Gordon equation driven by space-time white noise, we prove that the second-order increments of the solution can be approximated, after scaling with the diffusion coefficient, by those of the corresponding linearized stochastic Klein-Gordon equation. This extends the result of Huang et al. \cite{HOO2024} for the stochastic wave equation. A key difficulty arises from the more complex structure of the Green function, which we overcome by means of subtle analytical estimates. As applications, we analyze the quadratic variation of the solution and construct a consistent estimator for the diffusion parameter.

math.PR↗

RecruitScope: A Visual Analytics System for Multidimensional Recruitment Data Analysis

Online recruitment platforms have become the dominant channel for modern hiring, yet most platforms offer only basic filtering capabilities, such as job title, keyword, and salary range. This hinders comprehensive analysis of multi-attribute relationships and job market patterns across different scales. We present RecruitScope, a visual analytics system designed to support multidimensional and cross-level exploration of recruitment data for job seekers and employers, particularly HR specialists. Through coordinated visualizations, RecruitScope enables users to analyze job positions and salary patterns from multiple perspectives, interpret industry dynamics at the macro level, and identify emerging positions at the micro level. We demonstrate the effectiveness of RecruitScope through case studies that reveal regional salary distribution patterns, characterize industry growth trajectories, and discover high-demand emerging roles in the job market.

cs.HC↗

Kimina Lean Server: A High-Performance Lean Server for Large-Scale Verification

We introduce the Kimina Lean Server, an open-source project designed as a high-performance verifier for reinforcement learning pipelines. Built on top of the Lean REPL (Read-Eval-Print Loop) maintained by the Lean FRO, our server combines server-side parallelism by managing multiple Lean processes in parallel with a Least Recently Used (LRU) caching mechanism that reuses Lean imports across requests. On the client side, a lightweight Python package enables submitting proof batches and receiving Lean feedback, including extracted tactics and tactic states. Together, these features enable a scalable workflow for large-scale verification and data extraction. In our experiments, the Kimina Lean Server outperforms previous Lean interaction tools, achieving a 1.5 to 2 times speedup in verification time. Moreover, its improved efficiency has enabled its use in the large-scale training of state-of-the-art models such as Kimina-Prover. We hope that our open-source project will support the neural theorem proving community and accelerate future progress by enabling efficient large-scale verification and proof data extraction.

cs.LO↗

Conch: Competitive Debate Analysis via Visualizing Clash Points and Hierarchical Strategies

In-depth analysis of competitive debates is essential for participants to develop argumentative skills and refine strategies, and further improve their debating performance. However, manual analysis of unstructured and unlabeled textual records of debating is time-consuming and ineffective, as it is challenging to reconstruct contextual semantics and track logical connections from raw data. To address this, we propose Conch, an interactive visualization system that systematically analyzes both what is debated and how it is debated. In particular, we propose a novel parallel spiral visualization that compactly traces the multidimensional evolution of clash points and participant interactions throughout debate process. In addition, we leverage large language models with well-designed prompts to automatically identify critical debate elements such as clash points, disagreements, viewpoints, and strategies, enabling participants to understand the debate context comprehensively. Finally, through two case studies on real-world debates and a carefully-designed user study, we demonstrate Conch's effectiveness and usability for competitive debate analysis.

cs.HC↗

Object-IR: Leveraging Object Consistency and Mesh Deformation for Self-Supervised Image Retargeting

Eliminating geometric distortion in semantically important regions remains an intractable challenge in image retargeting. This paper presents Object-IR, a self-supervised architecture that reformulates image retargeting as a learning-based mesh warping optimization problem, where the mesh deformation is guided by object appearance consistency and geometric-preserving constraints. Given an input image and a target aspect ratio, we initialize a uniform rigid mesh at the output resolution and use a convolutional neural network to predict the motion of each mesh grid and obtain the deformed mesh. The retargeted result is generated by warping the input image according to the rigid mesh in the input image and the deformed mesh in the output resolution. To mitigate geometric distortion, we design a comprehensive objective function incorporating a) object-consistent loss to ensure that the important semantic objects retain their appearance, b) geometric-preserving loss to constrain simple scale transform of the important meshes, and c) boundary loss to enforce a clean rectangular output. Notably, our self-supervised paradigm eliminates the need for manually annotated retargeting datasets by deriving supervision directly from the input's geometric and semantic properties. Extensive evaluations on the RetargetMe benchmark demonstrate that our Object-IR achieves state-of-the-art performance, outperforming existing methods in quantitative metrics and subjective visual quality assessments. The framework efficiently processes arbitrary input resolutions (average inference time: 0.009s for 1024x683 resolution) while maintaining real-time performance on consumer-grade GPUs. The source code will soon be available at https://github.com/tlliao/Object-IR.

cs.CV↗

Local linearization for estimating the diffusion parameter of nonlinear stochastic wave equations with spatially correlated noise

We study the bi-parameter local linearization of the one-dimensional nonlinear stochastic wave equation driven by a Gaussian noise, which is white in time and has a spatially homogeneous covariance structure of Riesz-kernel type. We establish that the second-order increments of the solution can be approximated by those of the corresponding linearized wave equation, modulated by the diffusion coefficient. These findings extend the previous results of Huang et al. \cite{HOO2024}, which addressed the case of space-time white noise. As applications, we analyze the quadratic variation of the solution and construct a consistent estimator for the diffusion parameter.

math.PR↗