Showing 1–50 of 218 results for author: Chang, Q

CP Polarimetry with Linearly Polarized Photon Fusion and Double-Tagged Protons

Authors: Qi-Hui Chang, Shuai Zhao

Abstract: Double forward-proton tagging turns the forward detectors into event-by-event photon polarimeters because each measured proton recoil fixes the transverse momentum, and hence the linear-polarization axis, of the emitted photon. We show that this production-side polarimetry gives a decay-analyzer-independent measurement of the CP phase of a photon-coupled spin-zero resonance. We derive the leading-… ▽ More Double forward-proton tagging turns the forward detectors into event-by-event photon polarimeters because each measured proton recoil fixes the transverse momentum, and hence the linear-polarization axis, of the emitted photon. We show that this production-side polarimetry gives a decay-analyzer-independent measurement of the CP phase of a photon-coupled spin-zero resonance. We derive the leading-power photon-density contraction for a CP-mixed hard amplitude with scalar and pseudoscalar couplings. The CP phase appears as a translation of the second harmonic in the signed proton--proton azimuthal angle. This provides a compact production-side CP measurement for axionlike particles and more general spin-zero resonances. △ Less

Submitted 5 August, 2026; originally announced August 2026.

Comments: 7 pages, 2 figures

UXBench: Benchmarking User Experience in AI Assistants

Authors: Mengze Hong, Xia Zeng, Zeyang Lei, Sheng Wang, Chen Jason Zhang, Di Jiang, Taiming Fu, Jinfeng Huang, Mengqiao Liu, Qinghe Chang, Haosheng Zou, Qiongyi Zhou, Sijun He, Simonjmdeng, Haojing Huang, Zijian Li, Lucas Mu Li, Fubao Zhang, Mona Zhou, Wei Ma, Yuan Hua, Qi Zhu, Shuo Jiang, Chenxuan Ma, Yuanmeng Zhang , et al. (4 additional authors not shown)

Abstract: As AI assistants serve millions of users daily, evaluating user experience (UX) beyond general model capability has become increasingly important. We present UXBench, the first user-centric benchmark grounded in real user feedback signals for evaluating preference alignment and dialogue generation. The benchmark consists of three interconnected tasks, UX Judge, UX Eval, and UX Recovery, with 7,400… ▽ More As AI assistants serve millions of users daily, evaluating user experience (UX) beyond general model capability has become increasingly important. We present UXBench, the first user-centric benchmark grounded in real user feedback signals for evaluating preference alignment and dialogue generation. The benchmark consists of three interconnected tasks, UX Judge, UX Eval, and UX Recovery, with 7,400 test instances extracted from over 70K interaction logs of a mainstream Chinese AI assistant. The dataset closely reflects real user distributions, covering 8 scenarios, 83 domains, and diverse failure patterns that pose severe challenges. Extensive experiments on 26 frontier language models provide novel insights into how well models perceive user experience and how improvements in model capability contribute to better dialogue engagement. Through comprehensive analysis of model behavior and performance gaps, we show that user feedback prediction is a learnable capability, where a reward model trained from in-the-wild feedback signals can achieve well-calibrated accuracy. We further document the systematic biases of LLM-as-a-judge evaluation protocols and compare typical response strategies that directly affect user experience. UXBench establishes a new evaluation landscape and calls for greater attention to tailored UX optimization, contributing to a user-centric scaling law that shapes the success of AI assistants. △ Less

Submitted 14 July, 2026; v1 submitted 8 June, 2026; originally announced June 2026.

Asymptotic Optimality of Thompson Sampling for Risk-Averse Bandits with Sub-Gaussian Rewards

Authors: Joel Q. L. Chang

Abstract: We prove that $ρ\text{-}\mathrm{NPTS}_{\mathrm{SG}}$, an anchor-free nonparametric Thompson Sampling algorithm for risk-averse bandits, achieves regret matching the instance-dependent lower bound to leading order in $\log n$, establishing it as asymptotically optimal for any continuous risk functional $ρ$ (CVaR, mean-variance, Sharpe ratio, distortion risk measures, and more) on the class of distr… ▽ More We prove that $ρ\text{-}\mathrm{NPTS}_{\mathrm{SG}}$, an anchor-free nonparametric Thompson Sampling algorithm for risk-averse bandits, achieves regret matching the instance-dependent lower bound to leading order in $\log n$, establishing it as asymptotically optimal for any continuous risk functional $ρ$ (CVaR, mean-variance, Sharpe ratio, distortion risk measures, and more) on the class of distributions with bounded density and sub-Gaussian tails, including Gaussian arms. Both this result and its bounded-support counterpart require only continuity of $ρ$: strictly weaker than the dominance condition of prior parametric Thompson Sampling results, and strictly weaker than the Lipschitz condition of UCB-type algorithms, yielding the first instance-optimal guarantees for non-Lipschitz functionals such as the Sharpe ratio without parametric reward assumptions. The bounded-support case is developed first as a stepping stone sharing the same proof structure. The key technical contributions are a discretisation lemma (bounded support) and a truncated discretisation lemma (sub-Gaussian tails), each projecting the growing-alphabet Dirichlet posterior onto a fixed grid via the Dirichlet aggregation property, holding all polynomial prefactors at fixed degree independent of sample size and breaking the super-exponential barrier that blocked prior proofs. △ Less

Submitted 8 June, 2026; originally announced June 2026.

Comments: 10 pages, 4 figures

RadiusFPS: Efficient Farthest Point Sampling on CPUs and GPUs via Spherical Voxel Pruning

Authors: Ziyang Yu, Xiang Li, Qiong Chang, Jun Miyazaki

Abstract: Point clouds are a primary sensory representation for robotic perception, underpinning LiDAR-based autonomous driving, simultaneous localization and mapping (SLAM), and navigation. Within these pipelines, Farthest Point Sampling (FPS) is the most well-known downsampling operator, as its uniform coverage preserves the geometric structure on which downstream perception relies. However, the large tim… ▽ More Point clouds are a primary sensory representation for robotic perception, underpinning LiDAR-based autonomous driving, simultaneous localization and mapping (SLAM), and navigation. Within these pipelines, Farthest Point Sampling (FPS) is the most well-known downsampling operator, as its uniform coverage preserves the geometric structure on which downstream perception relies. However, the large time complexity of classical FPS scales poorly with the million-point-per-second rates of modern 3D sensors, making it a dominant latency bottleneck that conflicts with the real-time and limited onboard compute budgets of robotic systems. Therefore, we propose RadiusFPS, an FPS acceleration framework based on spherical voxel pruning that preserves the standard FPS update rule under the same initialization and tie-breaking policy. By indexing the point cloud with spherical voxels, RadiusFPS derives a conservative geometric bound that prunes redundant distance computations in each iteration, complemented by a coordinate-wise point-skip test that removes residual updates. We further introduce RadiusFPS-G, a warp-level GPU implementation that fuses voxel selection, pruning, and distance update into memory-coalesced kernels, eliminating costly global-memory round-trips. On indoor (S3DIS, ScanNet) and outdoor LiDAR (SemanticKITTI) benchmarks, RadiusFPS-G attains up to 2.5x speedup over GPU-based FPS and matches or exceeds QuickFPS among the evaluated methods while using roughly half its GPU memory, with comparable segmentation accuracy. When coupled with the learning-based FastPoint sampler, the resulting pipeline achieves the fastest End-to-End inference among all evaluated configurations. These properties make high-quality FPS-style sampling practical for latency- and memory-constrained robotic vision. △ Less

Submitted 4 June, 2026; originally announced June 2026.

Comments: 28 pages,15 figures

PEARL: Training Socratic Tutors with Pedagogically Aligned Reinforcement Learning

Authors: Qikai Chang, Zhenrong Zhang, Linbo Chen, Pengfei Hu, Jianshu Zhang, Youhui Guo, Jun Du

Abstract: Large Language Models (LLMs) have shown promise as educational tutors, yet effective tutoring requires more than solving problems: it must provide progressive Socratic guidance and balance multiple pedagogical objectives across multi-turn interactions. However, training such tutors remains challenging due to limited-fidelity and weakly controllable student simulation, under-specified pedagogical r… ▽ More Large Language Models (LLMs) have shown promise as educational tutors, yet effective tutoring requires more than solving problems: it must provide progressive Socratic guidance and balance multiple pedagogical objectives across multi-turn interactions. However, training such tutors remains challenging due to limited-fidelity and weakly controllable student simulation, under-specified pedagogical reward modeling, and unstable multi-objective optimization. To overcome these limitations, we propose PEARL, a pedagogically aligned reinforcement learning framework for training Socratic tutoring agents, consisting of three key components. First, we introduce a controllable student simulator that decouples latent cognitive states from response generation to model diverse abilities and misconceptions. Second, we develop a generative reward model that jointly evaluates pedagogical quality and objective correctness for policy optimization. Finally, we propose a stable multi-objective RL scheme that discretizes rewards within each dimension and aggregates normalized advantages across dimensions, preventing high-variance objectives from dominating updates. Experiments on multiple benchmarks show that PEARL achieves the best performance among open-source models and remains competitive with leading proprietary LLMs, despite using only a 30B policy model. △ Less

Submitted 28 May, 2026; originally announced May 2026.

Comments: 16 pages, 7 figures

Forget Less, Generalize More: Unifying Temporal and Structural Adaptation for Dynamic Graphs

Authors: Qian Chang, Ciprian Doru Giurcaneanu, Runsong Jia, Xia Li, Guoping Hu, Xiufeng Cheng, Jinqing Yang, Mengjia Wu, Yi Zhang

Abstract: Representation learning on dynamic graphs requires capturing complex dependencies that evolve across both time and structure. Existing approaches typically adopt fixed temporal decay schemes or predetermined structural propagation depths, limiting their ability to generalize across graphs with diverse interaction frequencies and topological characteristics. We propose Dual-Scale Retentive Dynamics… ▽ More Representation learning on dynamic graphs requires capturing complex dependencies that evolve across both time and structure. Existing approaches typically adopt fixed temporal decay schemes or predetermined structural propagation depths, limiting their ability to generalize across graphs with diverse interaction frequencies and topological characteristics. We propose Dual-Scale Retentive Dynamics (DSRD), a unified framework that maintains a retentive representation state encoding both temporal memory and structural context. DSRD introduces two key components: (i) a retentive state with dual-scale adaptation that jointly models temporal dynamics and structural propagation within a single recurrent formulation, and (ii) adaptive decay kernels with learnable time-sensitivity parameters that automatically balance short-term responsiveness and long-term retention based on the underlying interaction patterns. We provide theoretical analysis establishing the equivalence between event-wise parallel aggregation and efficient recurrent state updates, as well as stability and boundedness guarantees for the learned dynamics. Extensive experiments on 14 real-world benchmarks demonstrate that DSRD consistently achieves state-of-the-art performance on both link prediction and node classification tasks, with strong generalization across transductive and inductive settings. △ Less

Submitted 28 May, 2026; originally announced May 2026.

Semileptonic decay of $Λ_b^0 \to Λ_c (2860)^+/Λ_c(2625)^+\ell^-\overlineν_\ell$ within QCD light-cone sum rules

Authors: Hui-Hui Duan, Jia-Bao Feng, Feng-Mei Liu, Qin Chang

Abstract: In this work, we calculate the transition form factors for the weak decays $Λ_b^0 \to Λ_c(2860)^+$ and $Λ_b^0 \to Λ_c(2625)^+$ using QCD light-cone sum rules, and compute the branching fractions of the corresponding semileptonic decays $Λ_b^0 \to Λ_c(2860)^+ \ell^- \barν_\ell$ and $Λ_b^0 \to Λ_c(2625)^+ \ell^- \barν_\ell$. Our predicted branching fraction for… ▽ More In this work, we calculate the transition form factors for the weak decays $Λ_b^0 \to Λ_c(2860)^+$ and $Λ_b^0 \to Λ_c(2625)^+$ using QCD light-cone sum rules, and compute the branching fractions of the corresponding semileptonic decays $Λ_b^0 \to Λ_c(2860)^+ \ell^- \barν_\ell$ and $Λ_b^0 \to Λ_c(2625)^+ \ell^- \barν_\ell$. Our predicted branching fraction for $Λ_b^0 \to Λ_c(2625)^+ \ell^- \barν_\ell$ is consistent with experimental data and other theoretical predictions, validating the reliability of our method. On this basis, we also present the branching fraction of $Λ_b^0 \to Λ_c(2860)^+ \ell^- \barν_\ell$. These results may serve as a theoretical reference for future experimental measurements of this decay channel. △ Less

Submitted 22 May, 2026; originally announced May 2026.

Repurposing and Evaluating the (In)Feasibility of Dataset Poisoning enabled Watermarking for Contrastive Learning

Authors: Zhiyang Dai, Yansong Gao, Boyu Kuang, Haodong Li, Qi Chang, Gaurav Varshney, Derek Abbott, Anmin Fu

Abstract: Contrastive learning (CL) reduces annotation cost via auto-derived supervisory signals. Since large-scale in-house CL datasets are infeasible, reliance on third-party or internet data is common. Recent studies show CL models are vulnerable to data-poisoning backdoor attacks, but their generalization and robustness are underexplored. We systematically evaluate existing data-poisoning backdoor attac… ▽ More Contrastive learning (CL) reduces annotation cost via auto-derived supervisory signals. Since large-scale in-house CL datasets are infeasible, reliance on third-party or internet data is common. Recent studies show CL models are vulnerable to data-poisoning backdoor attacks, but their generalization and robustness are underexplored. We systematically evaluate existing data-poisoning backdoor attacks on CL, revealing limitations: poor dataset adaptability, low success rates, limited portability, and restrictive assumptions (e.g., downstream task knowledge). Interestingly, trigger samples exhibit distinguishable statistical divergence from clean samples, which inspires repurposing it as a watermark for dataset IP protection. Direct repurposing is challenging due to low success rates; we overcome this by statistical verification using a unified density metric. We further propose a multi-level watermarking scheme adapting to feature-level, soft-label, or hard-label outputs in CL. Experiments show some backdoor attacks can be repurposed as effective watermarks with trade-offs among fidelity, verifiability, and robustness. This work demonstrates weak backdoor effects become reliable signals for dataset IP protection in challenging CL settings. △ Less

Submitted 3 May, 2026; originally announced May 2026.

Holographic Schwinger Effect In a Step Dilaton Background

Authors: Sara Tahery, Qin Chang

Abstract: We investigate the holographic Schwinger effect in a confining background with a step dilaton profile, which induces a sharp transition between ultraviolet and infrared regimes and provides a qualitatively distinct realization of confinement. Within this framework, the quark--antiquark potential is extracted from the classical configuration of a fundamental string, allowing for a direct analysis o… ▽ More We investigate the holographic Schwinger effect in a confining background with a step dilaton profile, which induces a sharp transition between ultraviolet and infrared regimes and provides a qualitatively distinct realization of confinement. Within this framework, the quark--antiquark potential is extracted from the classical configuration of a fundamental string, allowing for a direct analysis of vacuum instability and pair production. In the absence of a magnetic field, the step dilaton leads to a significantly sharper suppression of the potential barrier as the electric field increases, implying an enhanced sensitivity of the critical electric field compared to smooth soft-wall models and demonstrating that the abrupt geometric transition qualitatively enhances the onset of vacuum decay. Incorporating an external magnetic field through the Dirac--Born--Infeld action, we find a nontrivial and amplified deformation of the potential barrier, resulting in a pronounced shift of the critical electric field that depends on both the magnitude and orientation of the magnetic field. Overall, the step dilaton background exhibits a substantially stronger response of the Schwinger effect to external electromagnetic fields than conventional soft-wall models, providing a novel mechanism for controlling pair production and highlighting the crucial role of dilaton structure in non-perturbative dynamics of holographic QCD. △ Less

Submitted 27 April, 2026; v1 submitted 19 April, 2026; originally announced April 2026.

Comments: 20 pages, 7 figures, references have been added

Journal ref: JHEP06(2026)181

Train-Small Deploy-Large: Leveraging Diffusion-Based Multi-Robot Planning

Authors: Siddharth Singh, Soumee Guha, Qing Chang, Scott Acton

Abstract: Learning based multi-robot path planning methods struggle to scale or generalize to changes, particularly variations in the number of robots during deployment. Most existing methods are trained on a fixed number of robots and may tolerate a reduced number during testing, but typically fail when the number increases. Additionally, training such methods for a larger number of agents can be both time… ▽ More Learning based multi-robot path planning methods struggle to scale or generalize to changes, particularly variations in the number of robots during deployment. Most existing methods are trained on a fixed number of robots and may tolerate a reduced number during testing, but typically fail when the number increases. Additionally, training such methods for a larger number of agents can be both time consuming and computationally expensive. However, analytical methods can struggle to scale computationally or handle dynamic changes in the environment. In this work, we propose to leverage a diffusion model based planner capable of handling dynamically varying number of agents. Our approach is trained on a limited number of agents and generalizes effectively to larger numbers of agents during deployment. Results show that integrating a single shared diffusion model based planner with dedicated inter-agent attention computation and temporal convolution enables a train small deploy-large paradigm with good accuracy. We validate our method across multiple scenarios and compare the performance with existing multi-agent reinforcement learning techniques and heuristic control based methods. △ Less

Submitted 7 April, 2026; originally announced April 2026.

Light-cone Distribution Amplitudes of Vector Mesons within the Self-Consistent Light-front Quark Model

Authors: Xiao-Nan Li, Shuai Xu, Qin Chang

Abstract: In this paper, we investigate the twist-2 and twist-3 light-cone distribution amplitudes (LCDAs) of vector mesons within the self-consistent light-front quark model, and the $n$th Gegenbauer moment $a_n(μ)$, $ξ$-moment $\langle ξ^n\rangle^{\parallel(\perp)}$ and transverse moment $\langle \mathbf{k}^n_\perp\rangle^{\parallel(\perp)}$ are also carried out. Adopting the parameter set fixed by confin… ▽ More In this paper, we investigate the twist-2 and twist-3 light-cone distribution amplitudes (LCDAs) of vector mesons within the self-consistent light-front quark model, and the $n$th Gegenbauer moment $a_n(μ)$, $ξ$-moment $\langle ξ^n\rangle^{\parallel(\perp)}$ and transverse moment $\langle \mathbf{k}^n_\perp\rangle^{\parallel(\perp)}$ are also carried out. Adopting the parameter set fixed by confinements from the mesonic decay constants, we perform the numerical analysis and the results reveal several key insights: (i) The flavor symmetry breaking effects are more pronounced in the twist-3 LCDAs of vector mesons, which leads to the establishment of $a_1^\perp>a_1^\parallel$ and $\langle ξ^n\rangle^\perp>\langle ξ^n\rangle^\parallel$. This is consistent with previous findings in research for the LCDAs of pseudoscalar mesons. (ii) For vector mesons, the twist dependence decreases in the heavy quark limit which lead to $φ_{2}^\parallel(x)\simeqφ_{3}^\perp(x)$. For pseudoscalar and vector mesons composed of the same quark constituents, their LCDAs with the same twist exhibit similarity and gradually converge as the increasing of quark mass, i.e., $φ_{2}^A(x)\simeqφ_{2}^\parallel(x)$ and $φ_{3}^P(x)\simeqφ_{3}^\perp(x)$ in $m_q\rightarrow\infty$. In the heavy-quark limit within the self-consistent LFQM framework, we find an approximate correlated spin-twist independence pattern, with $φ_2^A(x) \simeq φ_2^{\parallel}(x) \approxφ_3^P(x) \simeq φ_3^{\perp}(x)$, resulting from the suppression of twist dependence and the approximate spin independence between pseudoscalar and vector mesons. △ Less

Submitted 15 June, 2026; v1 submitted 22 March, 2026; originally announced March 2026.

Comments: 10 pages, 5 figures, Version accepted for publication in PLB

VLM-Guided Iterative Refinement for Surgical Image Segmentation with Foundation Models

Authors: Ange Lou, Yamin Li, Qi Chang, Nan Xi, Luyuan Xie, Zichao Li, Tianyu Luan

Abstract: Surgical image segmentation is essential for robot-assisted surgery and intraoperative guidance. However, existing methods are constrained to predefined categories, produce one-shot predictions without adaptive refinement, and lack mechanisms for clinician interaction. We propose IR-SIS, an iterative refinement system for surgical image segmentation that accepts natural language descriptions. IR-S… ▽ More Surgical image segmentation is essential for robot-assisted surgery and intraoperative guidance. However, existing methods are constrained to predefined categories, produce one-shot predictions without adaptive refinement, and lack mechanisms for clinician interaction. We propose IR-SIS, an iterative refinement system for surgical image segmentation that accepts natural language descriptions. IR-SIS leverages a fine-tuned SAM3 for initial segmentation, employs a Vision-Language Model to detect instruments and assess segmentation quality, and applies an agentic workflow that adaptively selects refinement strategies. The system supports clinician-in-the-loop interaction through natural language feedback. We also construct a multi-granularity language-annotated dataset from EndoVis2017 and EndoVis2018 benchmarks. Experiments demonstrate state-of-the-art performance on both in-domain and out-of-distribution data, with clinician interaction providing additional improvements. Our work establishes the first language-based surgical segmentation framework with adaptive self-refinement capabilities. △ Less

Submitted 9 February, 2026; originally announced February 2026.

Radiative generation of chiral vector operators in $b\to s ν\barν$ transition

Authors: Xin-Shuai Yan, Wen-Feng Liu, Qin Chang, Ya-Dong Yang

Abstract: The recent Belle II evidence for $B^+ \to K^+ ν\barν$, combined with a suppressed branching fraction ratio $R \equiv \mathcal{B}(B^0 \to K^{*0} ν\barν) / \mathcal{B}(B^+ \to K^+ ν\barν)$, necessitates new physics contributing to both left- and right-handed vector operators. We perform a systematic topological classification of one-loop completions that radiatively generate both operators without t… ▽ More The recent Belle II evidence for $B^+ \to K^+ ν\barν$, combined with a suppressed branching fraction ratio $R \equiv \mathcal{B}(B^0 \to K^{*0} ν\barν) / \mathcal{B}(B^+ \to K^+ ν\barν)$, necessitates new physics contributing to both left- and right-handed vector operators. We perform a systematic topological classification of one-loop completions that radiatively generate both operators without tree-level mediation, and construct two minimal benchmark scenarios: a scalar-rich model and a fermion-rich model. Evaluating these frameworks under specific benchmark mass schemes and four distinct flavor structures, we find a generic anti-correlation where enhancing one decay channel typically suppresses the other. A notable exception is a decoupled flavor configuration within the scalar-rich model, which yields simultaneous constructive interference, reducing $R$ by $4.22\%$ while satisfying all complementary flavor bounds. Under a universal coupling assumption, the scalar-rich model yields exact cancellation in the charged mode, highlighting the non-trivial interplay between flavor structure and loop-generated Wilson coefficients. △ Less

Submitted 17 June, 2026; v1 submitted 26 January, 2026; originally announced January 2026.

Comments: 27 pages, 9 figures, and 18 tables

Quarkonium light-cone distribution amplitudes: twist structure and mass dependence

Authors: Shuai Xu, Xiao-Nan Li, Jin-Zhong Han, Bai-Hui Cheng, Li-Li Chen, Qin Chang

Abstract: We present a systematic study of the leading- and next-to-leading-twist light-cone distribution amplitudes (LCDAs) of ground-state pseudoscalar and vector quarkonium within the light-front quark model (LFQM). By implementing the replacement $M \to M_0$, we analyze the longitudinal and transverse structures of the LCDAs, together with their Gegenbauer moments, $ξ$-moments, and transverse momentum m… ▽ More We present a systematic study of the leading- and next-to-leading-twist light-cone distribution amplitudes (LCDAs) of ground-state pseudoscalar and vector quarkonium within the light-front quark model (LFQM). By implementing the replacement $M \to M_0$, we analyze the longitudinal and transverse structures of the LCDAs, together with their Gegenbauer moments, $ξ$-moments, and transverse momentum moments. We show that charge-conjugation symmetry enforces the exact vanishing of all odd Gegenbauer moments and odd $ξ$-moments. For pseudoscalar quarkonium, the twist-2 and twist-3 LCDAs become identical, which leads to the same Gegenbauer moments, $ξ$-moments, and transverse momentum moments. For vector quarkonium, although the twist-2 and twist-3 LCDAs differ in the case of finite quark masses, they progressively converge as the quark mass increases. In the heavy-quark limit, all quarkonium LCDAs satisfy $φ^A_{2} = φ^P_{3} \simeq φ^{\parallel}_{2} \simeq φ^{\perp}_{3}$, demonstrating an emergent twist-independence of quarkonium distribution amplitudes. We further find that the LCDAs become increasingly peaked and narrower with increasing quark mass, indicating that the meson system becomes increasingly close to a nonrelativistic bound state, with a more uniform and stable distribution of internal longitudinal momentum. For all quarkonium, the peak value exhibits a simple phenomenological scaling behavior governed by the ratio $m/β$. The transverse momentum moments increase with the meson mass, indicating a progressively more compact bound-state structure. These results reveal a universal and systematic evolution of quarkonium LCDAs driven by the quark mass. △ Less

Submitted 20 March, 2026; v1 submitted 11 January, 2026; originally announced January 2026.

Comments: 18 pages, 3 figures . Version accepted for publication in JHEP

Probing Dark Matter annihilation in the Galactic Centre with TRIDENT

Authors: Yingwei Wang, Xinhui Chu, Andrew Cheek, Iwan Morton-Blake, Qichao Chang, Gwenael Giacinti, Samy Kaci, Xin Xiang, Donglian Xu, Fuyudi Zhang

Abstract: We determine the future sensitivity of the TRIDENT neutrino telescope to dark matter annihilation in the Galactic Centre. By applying the full detector design we show that TRIDENT will probe annihilation rates down to $\langleσv\rangle\approx5\times10^{-27}\,{\rm cm}^3\,{\rm s}^{-1}$ for a $10\,{\rm TeV}$ dark matter, which is below the thermal freeze-out benchmark. The analysis is carried out wit… ▽ More We determine the future sensitivity of the TRIDENT neutrino telescope to dark matter annihilation in the Galactic Centre. By applying the full detector design we show that TRIDENT will probe annihilation rates down to $\langleσv\rangle\approx5\times10^{-27}\,{\rm cm}^3\,{\rm s}^{-1}$ for a $10\,{\rm TeV}$ dark matter, which is below the thermal freeze-out benchmark. The analysis is carried out with all-flavour neutrino interactions, where we demonstrate that cascade events, primarily due to $ν_{e,τ}$, show greater sensitivity to a dark matter signal compared to the more commonly studied track events. Furthermore, we highlight the impact of a previously overlooked background, Galactic neutrinos produced from interactions between hadronic cosmic rays and interstellar gas. We find dark matter sensitivities are more strongly degraded in the high energy region above $\sim 10\, {\rm TeV}$, with a maximal weakening of approximately a factor of $\sim 2$. This effect remains smaller than the uncertainty associated with the dark matter density profile but can nonetheless mimic a positive annihilation signal. We contextualize these results with a concrete particle model and show that TRIDENT will be able to probe the most interesting untested parts of parameter space. △ Less

Submitted 11 January, 2026; originally announced January 2026.

Step Potential Advantage Estimation: Harnessing Intermediate Confidence and Correctness for Efficient Mathematical Reasoning

Authors: Fei Wu, Zhenrong Zhang, Qikai Chang, Jianshu Zhang, Quan Liu, Jun Du

Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) elicits long chain-of-thought reasoning in large language models (LLMs), but outcome-based rewards lead to coarse-grained advantage estimation. While existing approaches improve RLVR via token-level entropy or sequence-level length control, they lack a semantically grounded, step-level measure of reasoning progress. As a result, LLMs fail to di… ▽ More Reinforcement Learning with Verifiable Rewards (RLVR) elicits long chain-of-thought reasoning in large language models (LLMs), but outcome-based rewards lead to coarse-grained advantage estimation. While existing approaches improve RLVR via token-level entropy or sequence-level length control, they lack a semantically grounded, step-level measure of reasoning progress. As a result, LLMs fail to distinguish necessary deduction from redundant verification: they may continue checking after reaching a correct solution and, in extreme cases, overturn a correct trajectory into an incorrect final answer. To remedy the lack of process supervision, we introduce a training-free probing mechanism that extracts intermediate confidence and correctness and combines them into a Step Potential signal that explicitly estimates the reasoning state at each step. Building on this signal, we propose Step Potential Advantage Estimation (SPAE), a fine-grained credit assignment method that amplifies potential gains, penalizes potential drops, and applies penalty after potential saturates to encourage timely termination. Experiments across multiple benchmarks show SPAE consistently improves accuracy while substantially reducing response length, outperforming strong RL baselines and recent efficient reasoning and token-level advantage estimation methods. The code is available at https://github.com/cii030/SPAE-RL. △ Less

Submitted 7 January, 2026; originally announced January 2026.

Distribution Amplitudes of Pseudoscalar Mesons within the Light-Front Quark Model

Authors: Xiao-Nan Li, Shuai Xu, Qin Chang

Abstract: In this paper, we investigate the distribution amplitudes (DAs) of pseudoscalar mesons within the light-front quark model (LFQM), and the $n$-th Gegenbauer moment $a_n(μ)$, $ξ$-moment $\langle ξ^n\rangle$ and transverse moment $\langle \mathbf{k}^n_\perp\rangle$ are also studied. Two parameter sets are adopted by fitting to mesonic decay constants or mass spectra, leading to distinct schemes for n… ▽ More In this paper, we investigate the distribution amplitudes (DAs) of pseudoscalar mesons within the light-front quark model (LFQM), and the $n$-th Gegenbauer moment $a_n(μ)$, $ξ$-moment $\langle ξ^n\rangle$ and transverse moment $\langle \mathbf{k}^n_\perp\rangle$ are also studied. Two parameter sets are adopted by fitting to mesonic decay constants or mass spectra, leading to distinct schemes for numerical analysis. Our results reveal several key insights: (1) Under scheme-I, the second Gegenbauer moment for the pion, $a_{2,π} = 0.054$, is in excellent agreement with predictions from the platykurtic model and the nonlocal chiral quark model (NL$χ$QM). Meanwhile, scheme-II yields $a_{2,π} = 0.126$, which is consistent with lattice QCD (LQCD), Dyson-Schwinger equation (DSE) and data-driven light-cone sum rules (LCSR) analyses. Furthermore, we observe that $φ_{2,π} = φ^P_{3,π}$ with the replacement $M \to M_0$. (2) Flavor symmetry breaking effects are found to be more significant in higher-twist DAs, with a notable relation $a_1^P\approx 2 a_1$ emerging in heavy meson systems. (3) Interestingly, numerical analysis indicates that the transverse moment $\langle \mathbf{k}^n_\perp\rangle$ is twist-independent with the replacement $M\to M_0$. Additionally, we propose an empirical scaling relation $\frac{Δm^{0.61}}{β^{1.10}}x_p=0.67$ to effectively describe the shape of DAs for heavy mesons. Collectively, these results demonstrate that mass asymmetry remarkably influences DAs and kinds of moments, these theoretical predictions will be tested in future experiments. △ Less

Submitted 2 January, 2026; originally announced January 2026.

Comments: 18 pages, 4 figures

Search for Light Dark Matter in Rare Meson Decays

Authors: Ze-Kun Liu, Ying Li, Biao-Feng Hou, Qin Chang

Abstract: Current dark matter direct detection experiments have low sensitivity to sub-GeV dark matter. In this work, we demonstrate that rare $B$ and $K$ meson decays with missing energy in the final state can serve as efficient probes in this mass range. We analyze a generic $Z^{\prime}$ portal dark matter model and derive upper limits on its parameters from experimental bounds on the rare $B$ and $K$ mes… ▽ More Current dark matter direct detection experiments have low sensitivity to sub-GeV dark matter. In this work, we demonstrate that rare $B$ and $K$ meson decays with missing energy in the final state can serve as efficient probes in this mass range. We analyze a generic $Z^{\prime}$ portal dark matter model and derive upper limits on its parameters from experimental bounds on the rare $B$ and $K$ meson decays. Our results show that such meson decay processes provide complementary constraints to current direct detection experiments for sub-GeV dark matter, particularly for interaction forms mediated by dark matter momentum-dependent operators. △ Less

Submitted 2 January, 2026; v1 submitted 24 December, 2025; originally announced December 2025.

Comments: 18 pages, 5 figures, 2 tables

Searching for Ultralight Dark Matter with M{ö}ssbauer Resonance

Authors: Peng-Long Zhang, Yu-Ming Yang, Xiao-Jun Bi, Qin Chang, Yu Gao, Hai-Bo Li, Wei Xu, Peng-Fei Yin

Abstract: We investigate the feasibility of probing the interactions between ultralight scalar dark matter and atomic nuclei using a stationary Mössbauer spectroscopy scheme. The exceptional energy resolution of the Mössbauer resonance enables searches for tiny nuclear energy shifts induced by the local dark matter field. The dark matter mass range considered in this work is $10^{-18}$--… ▽ More We investigate the feasibility of probing the interactions between ultralight scalar dark matter and atomic nuclei using a stationary Mössbauer spectroscopy scheme. The exceptional energy resolution of the Mössbauer resonance enables searches for tiny nuclear energy shifts induced by the local dark matter field. The dark matter mass range considered in this work is $10^{-18}$--$10^{-8}~\mathrm{eV}$. We present projected constraints for two candidate Mössbauer isotopes, $^{109}\mathrm{Ag}$ and $^{45}\mathrm{Sc}$, with $^{109}\mathrm{Ag}$ providing the strongest sensitivity. For $^{109}\mathrm{Ag}$, projected sensitivities as low as approximately $10^{-19}$, $10^{-22}$, and $10^{-21}~\mathrm{GeV^{-1}}$ can be achieved for the scalar DM--photon, DM--gluon, and DM--quark couplings $f_γ^{-1}$, $f_{g}^{-1}$, and $f_{\hat{m}}^{-1}$, respectively. In the low-mass region, the projected sensitivity to the scalar DM--photon coupling approaches the current constraints from equivalence-principle (EP) tests. These results demonstrate that Mössbauer-based techniques provide a promising and competitive approach for probing ultralight dark matter interactions with Standard Model particles. △ Less

Submitted 2 August, 2026; v1 submitted 21 December, 2025; originally announced December 2025.

A dynamic memory assignment strategy for dilation-based ICP algorithm on embedded GPUs

Authors: Qiong Chang, Weimin Wang, Junpei Zhong, Jun Miyazaki

Abstract: This paper proposes a memory-efficient optimization strategy for the high-performance point cloud registration algorithm VANICP, enabling lightweight execution on embedded GPUs with constrained hardware resources. VANICP is a recently published acceleration framework that significantly improves the computational efficiency of point-cloud-based applications. By transforming the global nearest neigh… ▽ More This paper proposes a memory-efficient optimization strategy for the high-performance point cloud registration algorithm VANICP, enabling lightweight execution on embedded GPUs with constrained hardware resources. VANICP is a recently published acceleration framework that significantly improves the computational efficiency of point-cloud-based applications. By transforming the global nearest neighbor search into a localized process through a dilation-based information propagation mechanism, VANICP greatly reduces the computational complexity of the NNS. However, its original implementation demands a considerable amount of memory, which restricts its deployment in resource-constrained environments such as embedded systems. To address this issue, we propose a GPU-oriented dynamic memory assignment strategy that optimizes the memory usage of the dilation operation. Furthermore, based on this strategy, we construct an enhanced version of the VANICP framework that achieves over 97% reduction in memory consumption while preserving the original performance. Source code is published on: https://github.com/changqiong/VANICP4Em.git. △ Less

Submitted 4 December, 2025; originally announced December 2025.

GazeInterpreter: Parsing Eye Gaze to Generate Eye-Body-Coordinated Narrations

Authors: Qing Chang, Zhiming Hu

Abstract: Comprehensively interpreting human behavior is a core challenge in human-aware artificial intelligence. However, prior works typically focused on body behavior, neglecting the crucial role of eye gaze and its synergy with body motion. We present GazeInterpreter - a novel large language model-based (LLM-based) approach that parses eye gaze data to generate eye-body-coordinated narrations. Specifica… ▽ More Comprehensively interpreting human behavior is a core challenge in human-aware artificial intelligence. However, prior works typically focused on body behavior, neglecting the crucial role of eye gaze and its synergy with body motion. We present GazeInterpreter - a novel large language model-based (LLM-based) approach that parses eye gaze data to generate eye-body-coordinated narrations. Specifically, our method features 1) a symbolic gaze parser that translates raw gaze signals into symbolic gaze events; 2) a hierarchical structure that first uses an LLM to generate eye gaze narration at semantic level and then integrates gaze with body motion within the same observation window to produce integrated narration; and 3) a self-correcting loop that iteratively refines the modality match, temporal coherence, and completeness of the integrated narration. This hierarchical and iterative processing can effectively align physical values and semantic text in the temporal and spatial domains. We validated the effectiveness of our eye-body-coordinated narrations on the text-driven motion generation task in the large-scale Nymeria benchmark. Moreover, we report significant performance improvements for the sample downstream tasks of action anticipation and behavior summarization. Taken together, these results reveal the significant potential of parsing eye gaze to interpret human behavior and open up a new direction for human behavior understanding. △ Less

Submitted 20 November, 2025; originally announced November 2025.

Comments: Accepted to AAAI 2026. 9 pages, 4 figures

Optimising Underwater Neutrino Telescopes for All-Flavour Point Source Sensitivity

Authors: Iwan Morton-Blake, Fuyudi Zhang, Qichao Chang, Shuhua Hao, Weilun Huang, Hualin Mei, Wei Tian, Yingwei Wang, Xin Xiang, Donglian Xu

Abstract: High-energy neutrino astronomy has advanced rapidly in recent years, with IceCube, KM3NeT, and Baikal-GVD establishing a diffuse astrophysical flux and pointing to promising source candidates. These achievements mark the transition from first detections to detailed source studies, motivating next-generation detectors with larger volumes, improved angular resolution, and full neutrino-flavour sensi… ▽ More High-energy neutrino astronomy has advanced rapidly in recent years, with IceCube, KM3NeT, and Baikal-GVD establishing a diffuse astrophysical flux and pointing to promising source candidates. These achievements mark the transition from first detections to detailed source studies, motivating next-generation detectors with larger volumes, improved angular resolution, and full neutrino-flavour sensitivity. We present a performance study of large underwater neutrino telescopes, taking the proposed TRIDENT array in the South China Sea as a case study, with a focus on comparing the performance of various detector configurations against the TRIDENT baseline design. Both track-like events primarily from muon neutrinos, which provide precise directional information, and cascade events from all flavours, which offer superior energy resolution, diffuse-source sensitivity, and all-sky flavour coverage, are included to achieve a balanced performance across source types. The time to discover potential astrophysical sources with both track- and cascade-like events is used as the figure of merit to compare a variety of detector design choices. Our results show that, for a fixed number of optical modules, simply enlarging the instrumented volume does not inherently lead to improved performance, while taller strings can provide modest gains across all detector channels, within engineering constraints. Distributing dense clusters of strings over a large volume is found to generally worsen discovery potential compared to the baseline layout. Finally, the optical properties of the sea-water emerge as the key factor dictating the optimisation of detector layout, highlighting the need for in-situ measurements and early deployment of optical modules to guide the final array configuration. △ Less

Submitted 28 October, 2025; originally announced October 2025.

Comments: 16 pages, 7 figures

GRNND: A GPU-Parallel Relative NN-Descent Algorithm for Efficient Approximate Nearest Neighbor Graph Construction

Authors: Xiang Li, Qiong Chang, Yun Li, Jun Miyazaki

Abstract: Relative Nearest Neighbor Descent (RNN-Descent) is a state-of-the-art algorithm for constructing sparse approximate nearest neighbor (ANN) graphs by combining the iterative refinement of NN-Descent with the edge-pruning rules of the Relative Neighborhood Graph (RNG). It has demonstrated strong effectiveness in large-scale search tasks such as information retrieval and related tasks. However, as th… ▽ More Relative Nearest Neighbor Descent (RNN-Descent) is a state-of-the-art algorithm for constructing sparse approximate nearest neighbor (ANN) graphs by combining the iterative refinement of NN-Descent with the edge-pruning rules of the Relative Neighborhood Graph (RNG). It has demonstrated strong effectiveness in large-scale search tasks such as information retrieval and related tasks. However, as the amount and dimensionality of data increase, the complexity of graph construction in RNN-Descent rises sharply, making this stage increasingly time-consuming and even prohibitive for subsequent query processing. In this paper, we propose GRNND, the first GPU-parallel algorithm of RNN-Descent designed to fully exploit GPU architecture. GRNND introduces a disordered neighbor propagation strategy to mitigate synchronized update traps, enhancing structural diversity, and avoiding premature convergence during parallel execution. It also leverages warp-level cooperative operations and a double-buffered neighbor pool with fixed capacity for efficient memory access, eliminate contention, and enable highly parallelized neighbor updates. Extensive experiments demonstrate that GRNND consistently outperforms existing CPU- and GPU-based methods. GRNND achieves 2.4 to 51.7x speedup over existing GPU methods, and 17.8 to 49.8x speedup over CPU methods. △ Less

Submitted 3 October, 2025; originally announced October 2025.

Interpreting the 95 GeV di-photon and di-tau Excesses in the Georgi-Machacek Model

Authors: Qin Chang, Xiaokang Du, Pengxuan Zhu

Abstract: We revisit the 95~GeV $γγ$ and $ττ$ excesses in the Georgi--Machacek model using a combined fit to a single light $CP$-even custodial singlet $H$. Using the one-loop renormalization group-improved effective potential and positive-definiteness conditions for vacuum stability, together with perturbative unitarity, electroweak precision tests, $B$-physics, and Higgs data, we identify a narrow but via… ▽ More We revisit the 95~GeV $γγ$ and $ττ$ excesses in the Georgi--Machacek model using a combined fit to a single light $CP$-even custodial singlet $H$. Using the one-loop renormalization group-improved effective potential and positive-definiteness conditions for vacuum stability, together with perturbative unitarity, electroweak precision tests, $B$-physics, and Higgs data, we identify a narrow but viable parameter region. In the allowed region $μ_{γγ}$ approaches the experimental central value, while $μ_{ττ}$ reaches only $\sim0.5$ (about $1.4σ$ below the CMS central value) and remains within the $2σ$ interval - a good description of the di-photon excess but only a partial accommodation of the di-tau excess. The scenario predicts characteristic patterns in the singlet mixing and triplet vacuum expectation values and is highly testable at the HL-LHC and future lepton colliders via precision $κ_V$ measurements and direct exotic searches. △ Less

Submitted 19 August, 2026; v1 submitted 30 September, 2025; originally announced September 2025.

Comments: 31 pages, 6 figures, 1 table. published version

THOR: Tool-Integrated Hierarchical Optimization via RL for Mathematical Reasoning

Authors: Qikai Chang, Zhenrong Zhang, Pengfei Hu, Jun Du, Jiefeng Ma, Yicheng Pan, Jianshu Zhang, Quan Liu, Jianqing Gao

Abstract: Large Language Models (LLMs) have made remarkable progress in mathematical reasoning, but still continue to struggle with high-precision tasks like numerical computation and formal symbolic manipulation. Integrating external tools has emerged as a promising approach to bridge this gap. Despite recent advances, existing methods struggle with three key challenges: constructing tool-integrated reason… ▽ More Large Language Models (LLMs) have made remarkable progress in mathematical reasoning, but still continue to struggle with high-precision tasks like numerical computation and formal symbolic manipulation. Integrating external tools has emerged as a promising approach to bridge this gap. Despite recent advances, existing methods struggle with three key challenges: constructing tool-integrated reasoning data, performing fine-grained optimization, and enhancing inference. To overcome these limitations, we propose THOR (Tool-Integrated Hierarchical Optimization via RL). First, we introduce TIRGen, a multi-agent based pipeline for constructing high-quality datasets of tool-integrated reasoning paths, aligning with the policy and generalizing well across diverse models. Second, to perform fine-grained hierarchical optimization, we introduce an RL strategy that jointly optimizes for both episode-level problem solving and step-level code generation. This is motivated by our key insight that the success of an intermediate tool call is a strong predictor of the final answer's correctness. Finally, THOR incorporates a self-correction mechanism that leverages immediate tool feedback to dynamically revise erroneous reasoning paths during inference. Our approach demonstrates strong generalization across diverse models, performing effectively in both reasoning and non-reasoning models. It further achieves state-of-the-art performance for models of a similar scale on multiple mathematical benchmarks, while also delivering consistent improvements on code benchmarks. Our code will be publicly available at https://github.com/JingMog/THOR. △ Less

Submitted 10 February, 2026; v1 submitted 17 September, 2025; originally announced September 2025.

Comments: 22 pages, 13 figures, ICLR 2026

Mineral Detection of Neutrinos and Dark Matter 2025 Proceedings

Authors: Shigenobu Hirose, Patrick Stengel, Natsue Abe, Daniel Ang, Lorenzo Apollonio, Gabriela R. Araujo, Yoshihiro Asahara, Laura Baudis, Pranshu Bhaumik, Nathaniel Bowden, Joseph Bramante, Lorenzo Caccianiga, Mason Camp, Qing Chang, Jordan Chapman, Reza Ebadi, Alexey Elykov, Anna Erickson, Valentin Fondement, Katherine Freese, Shota Futamura, Claudio Galelli, Andrew Gilpin, Takeshi Hanyu, Noriko Hasebe , et al. (48 additional authors not shown)

Abstract: The third ``Mineral Detection of Neutrinos and Dark Matter'' (MD$ν$DM'25) meeting was held May 20-23, 2025 in Yokohama, Japan, hosted by the Yokohama Institute for Earth Sciences, Japan Agency for Marine-Earth Science and Technology (JAMSTEC). These proceedings compile contributions from the workshop and update the progress of mineral detector research. MD$ν$DM'25 was the third such meeting, follo… ▽ More The third ``Mineral Detection of Neutrinos and Dark Matter'' (MD$ν$DM'25) meeting was held May 20-23, 2025 in Yokohama, Japan, hosted by the Yokohama Institute for Earth Sciences, Japan Agency for Marine-Earth Science and Technology (JAMSTEC). These proceedings compile contributions from the workshop and update the progress of mineral detector research. MD$ν$DM'25 was the third such meeting, following the first in October of 2022 held at the IFPU in Trieste, Italy and the second in January of 2024 hosted by the Center for Neutrino Physics at Virginia Tech in Arlington, USA. Mineral detectors record and retain damage induced by nuclear recoils in synthetic or natural mineral samples. The damage features can then be read out by a variety of nano- and micro-scale imaging techniques. Applications of mineral detectors on timescales relevant for laboratory experiments include reactor neutrino monitoring and dark matter detection, with the potential to measure the directions as well as the energies of the induced nuclear recoils. For natural mineral detectors which record nuclear recoils over geological timescales, reading out even small mineral samples could be sensitive to rare interactions induced by astrophysical neutrinos, cosmic rays, dark matter and heavy exotic particles. A series of mineral detectors of different ages could measure the time evolution of these fluxes, offering a unique window into the history of our solar system and the Milky Way. Mineral detector research is highly multidisciplinary, incorporating aspects of high energy physics, condensed matter physics, materials science, geoscience, and AI/ML for data analysis. Although realizing the scientific potential of mineral detectors poses many challenges, the MD$ν$DM community looks forward to the continued development of mineral detector experiments and the possible discoveries that mineral detectors could reveal. △ Less

Submitted 28 August, 2025; originally announced August 2025.

Comments: Summary and proceedings of the MDvDM'25 conference, May 20-23 2025 in Yokohama, Japan

MCP-Bench: Benchmarking Tool-Using LLM Agents with Complex Real-World Tasks via MCP Servers

Authors: Zhenting Wang, Qi Chang, Hemani Patel, Shashank Biju, Cheng-En Wu, Quan Liu, Aolin Ding, Alireza Rezazadeh, Ankit Shah, Yujia Bao, Eugene Siow

Abstract: We introduce MCP-Bench, a benchmark for evaluating large language models (LLMs) on realistic, multi-step tasks that demand tool use, cross-tool coordination, precise parameter control, and planning/reasoning for solving tasks. Built on the Model Context Protocol (MCP), MCP-Bench connects LLMs to 28 representative live MCP servers spanning 250 tools across domains such as finance, traveling, scient… ▽ More We introduce MCP-Bench, a benchmark for evaluating large language models (LLMs) on realistic, multi-step tasks that demand tool use, cross-tool coordination, precise parameter control, and planning/reasoning for solving tasks. Built on the Model Context Protocol (MCP), MCP-Bench connects LLMs to 28 representative live MCP servers spanning 250 tools across domains such as finance, traveling, scientific computing, and academic search. Unlike prior API-based benchmarks, each MCP server provides a set of complementary tools designed to work together, enabling the construction of authentic, multi-step tasks with rich input-output coupling. Tasks in MCP-Bench test agents' ability to retrieve relevant tools from fuzzy instructions without explicit tool names, plan multi-hop execution trajectories for complex objectives, ground responses in intermediate tool outputs, and orchestrate cross-domain workflows - capabilities not adequately evaluated by existing benchmarks that rely on explicit tool specifications, shallow few-step workflows, and isolated domain operations. We propose a multi-faceted evaluation framework covering tool-level schema understanding and usage, trajectory-level planning, and task completion. Experiments on 20 advanced LLMs reveal persistent challenges in MCP-Bench. Code and data: https://github.com/Accenture/mcp-bench. △ Less

Submitted 28 August, 2025; originally announced August 2025.

The $Λ_{b} \to Λ$ transition form factors in perturbative QCD approach

Authors: Lei Yang, Jia-Jie Han, Qin Chang, Fu-Sheng Yu

Abstract: In this work, we investigate the $Λ_b \to Λ$ transition form factors in the perturbative QCD (PQCD) approach, incorporating higher-twist light-cone distribution amplitudes (LCDAs). The resulted form factors show that higher-twist LCDAs are dominant numerically. By combining our PQCD predictions at low-$q^2$ with lattice QCD results at high-$q^2$, $z$-series expansion fits are performed to parametr… ▽ More In this work, we investigate the $Λ_b \to Λ$ transition form factors in the perturbative QCD (PQCD) approach, incorporating higher-twist light-cone distribution amplitudes (LCDAs). The resulted form factors show that higher-twist LCDAs are dominant numerically. By combining our PQCD predictions at low-$q^2$ with lattice QCD results at high-$q^2$, $z$-series expansion fits are performed to parametrize the form factors over the full kinematic range. We also provide the prediction for physical observables in the rare decay $Λ_b \to Λμ^+ μ^-$, including the differential branching fraction, dilepton longitudinal polarization fraction, and forward-backward asymmetries (lepton-side, hadron-side, and combined lepton-hadron). Our obtained form factors are consistent with those in other theoretical methods within the uncertainties. △ Less

Submitted 1 February, 2026; v1 submitted 25 August, 2025; originally announced August 2025.

D$^2$-LIO: Enhanced Optimization for LiDAR-IMU Odometry Considering Directional Degeneracy

Authors: Guodong Yao, Hao Wang, Qing Chang

Abstract: LiDAR-inertial odometry (LIO) plays a vital role in achieving accurate localization and mapping, especially in complex environments. However, the presence of LiDAR feature degeneracy poses a major challenge to reliable state estimation. To overcome this issue, we propose an enhanced LIO framework that integrates adaptive outlier-tolerant correspondence with a scan-to-submap registration strategy.… ▽ More LiDAR-inertial odometry (LIO) plays a vital role in achieving accurate localization and mapping, especially in complex environments. However, the presence of LiDAR feature degeneracy poses a major challenge to reliable state estimation. To overcome this issue, we propose an enhanced LIO framework that integrates adaptive outlier-tolerant correspondence with a scan-to-submap registration strategy. The core contribution lies in an adaptive outlier removal threshold, which dynamically adjusts based on point-to-sensor distance and the motion amplitude of platform. This mechanism improves the robustness of feature matching in varying conditions. Moreover, we introduce a flexible scan-to-submap registration method that leverages IMU data to refine pose estimation, particularly in degenerate geometric configurations. To further enhance localization accuracy, we design a novel weighting matrix that fuses IMU preintegration covariance with a degeneration metric derived from the scan-to-submap process. Extensive experiments conducted in both indoor and outdoor environments-characterized by sparse or degenerate features-demonstrate that our method consistently outperforms state-of-the-art approaches in terms of both robustness and accuracy. △ Less

Submitted 19 August, 2025; originally announced August 2025.

Comments: 7 page, 2 figures

DOD-SA: Infrared-Visible Decoupled Object Detection with Single-Modality Annotations

Authors: Hang Jin, Chenqiang Gao, Junjie Guo, Fangcen Liu, Kanghui Tian, Qinyao Chang

Abstract: Infrared-visible object detection has shown great potential in real-world applications, enabling robust all-day perception by leveraging the complementary information of infrared and visible images. However, existing methods typically require dual-modality annotations to output detection results for both modalities during prediction, which incurs high annotation costs. To address this challenge, w… ▽ More Infrared-visible object detection has shown great potential in real-world applications, enabling robust all-day perception by leveraging the complementary information of infrared and visible images. However, existing methods typically require dual-modality annotations to output detection results for both modalities during prediction, which incurs high annotation costs. To address this challenge, we propose a novel infrared-visible Decoupled Object Detection framework with Single-modality Annotations, called DOD-SA. The architecture of DOD-SA is built upon a Single- and Dual-Modality Collaborative Teacher-Student Network (CoSD-TSNet), which consists of a single-modality branch (SM-Branch) and a dual-modality decoupled branch (DMD-Branch). The teacher model generates pseudo-labels for the unlabeled modality, simultaneously supporting the training of the student model. The collaborative design enables cross-modality knowledge transfer from the labeled modality to the unlabeled modality, and facilitates effective SM-to-DMD branch supervision. To further improve the decoupling ability of the model and the pseudo-label quality, we introduce a Progressive and Self-Tuning Training Strategy (PaST) that trains the model in three stages: (1) pretraining SM-Branch, (2) guiding the learning of DMD-Branch by SM-Branch, and (3) refining DMD-Branch. In addition, we design a Pseudo Label Assigner (PLA) to align and pair labels across modalities, explicitly addressing modality misalignment during training. Extensive experiments on the DroneVehicle dataset demonstrate that our method outperforms state-of-the-art (SOTA). △ Less

Submitted 14 August, 2025; originally announced August 2025.

Comments: 9 pages, 5 figures

MSC Class: 68T07; 68T45; 68U10 ACM Class: I.2.10

A Cost Effective Optimization of the hybrid-DOM Design for TRIDENT

Authors: Hengbin Shao, Fuyudi Zhang, Qichao Chang, Shuhua Hao, Ruike Cao, Jingtao Huang, Weilun Huang, Hai Liu, Hualin Mei, Iwan Morton-Blake, Wei Tian, Yingwei Wang, Xin Xiang, Donglian Xu

Abstract: TRIDENT is a planned multi-cubic-kilometer deep-sea neutrino telescope to be built in the South China Sea, designed to rapidly discover high-energy astrophysical neutrino sources with sensitivity to all neutrino flavors. Achieving this at scale requires a detector design that balances performance with power, cost, and mechanical simplicity. This study presents a cost-effective optimization of TRID… ▽ More TRIDENT is a planned multi-cubic-kilometer deep-sea neutrino telescope to be built in the South China Sea, designed to rapidly discover high-energy astrophysical neutrino sources with sensitivity to all neutrino flavors. Achieving this at scale requires a detector design that balances performance with power, cost, and mechanical simplicity. This study presents a cost-effective optimization of TRIDENT's hybrid Digital Optical Module (hDOM) design, comparing configurations using high-quantum-efficiency (QE) 3-inch PMTs and larger 4-inch PMTs, the latter evaluated with both baseline and enhanced QE assumptions. Using full-chain detector simulations incorporating site-specific seawater optical properties and realistic backgrounds, we assess performance in all-flavor neutrino detection efficiency, directional reconstruction, and tau neutrino flavor identification from 1 TeV to 10 PeV. We find that if 4-inch PMTs can achieve QE comparable to 3-inch PMTs, their performance matches or improves upon that of the 3-inch design, while significantly reducing channel count, power consumption, and cost. These findings support the 4-inch PMT hDOM as a promising and scalable choice for TRIDENT's future instrumentation. △ Less

Submitted 14 July, 2025; originally announced July 2025.

Comments: 15 pages, 7 figures

Polarization Probes of New Physics in Lepton-Flavor-Violating Hyperon Production from $e^- N \to τ^- Y$ Scattering

Authors: Xin-Shuai Yan, Zhen-Qing Hu, Qin Chang, Ya-Dong Yang

Abstract: While charged lepton-flavor-violating tau decays to strange mesons provide powerful probes of the underlying $τ\to\ell d \bar{s}$ transition, the corresponding decay modes involving hyperons are kinematically forbidden. To address this gap, we propose the quasi-elastic scattering processes $e^- N \to τ^- Y$. Within a general low-energy effective Lagrangian, we perform a comprehensive analysis of t… ▽ More While charged lepton-flavor-violating tau decays to strange mesons provide powerful probes of the underlying $τ\to\ell d \bar{s}$ transition, the corresponding decay modes involving hyperons are kinematically forbidden. To address this gap, we propose the quasi-elastic scattering processes $e^- N \to τ^- Y$. Within a general low-energy effective Lagrangian, we perform a comprehensive analysis of the final-state polarizations of both the $τ$ lepton and the hyperon $Y$, systematically addressing the theoretical uncertainty driven by the model dependence of form factors. Our analysis demonstrates that while this uncertainty leads to large variations in predicted rates, the polarization observables can potentially distinguish between the form factor models, and help to determine the chiral nature of the vectorial new physics interaction. Finally, we forecast the event rates for future facilities, revealing a strong dependence on the form factor models, with predicted yields ranging from potentially observable levels to rates far below current detection thresholds. △ Less

Submitted 9 July, 2025; v1 submitted 30 June, 2025; originally announced June 2025.

Comments: 18 pages, 9 figures, and 8 tables. Suggestions and comments are welcome

Faster than Fast: Accelerating Oriented FAST Feature Detection on Low-end Embedded GPUs

Authors: Qiong Chang, Xinyuan Chen, Xiang Li, Weimin Wang, Jun Miyazaki

Abstract: The visual-based SLAM (Simultaneous Localization and Mapping) is a technology widely used in applications such as robotic navigation and virtual reality, which primarily focuses on detecting feature points from visual images to construct an unknown environmental map and simultaneously determines its own location. It usually imposes stringent requirements on hardware power consumption, processing s… ▽ More The visual-based SLAM (Simultaneous Localization and Mapping) is a technology widely used in applications such as robotic navigation and virtual reality, which primarily focuses on detecting feature points from visual images to construct an unknown environmental map and simultaneously determines its own location. It usually imposes stringent requirements on hardware power consumption, processing speed and accuracy. Currently, the ORB (Oriented FAST and Rotated BRIEF)-based SLAM systems have exhibited superior performance in terms of processing speed and robustness. However, they still fall short of meeting the demands for real-time processing on mobile platforms. This limitation is primarily due to the time-consuming Oriented FAST calculations accounting for approximately half of the entire SLAM system. This paper presents two methods to accelerate the Oriented FAST feature detection on low-end embedded GPUs. These methods optimize the most time-consuming steps in Oriented FAST feature detection: FAST feature point detection and Harris corner detection, which is achieved by implementing a binary-level encoding strategy to determine candidate points quickly and a separable Harris detection strategy with efficient low-level GPU hardware-specific instructions. Extensive experiments on a Jetson TX2 embedded GPU demonstrate an average speedup of over 7.3 times compared to widely used OpenCV with GPU support. This significant improvement highlights its effectiveness and potential for real-time applications in mobile and resource-constrained environments. △ Less

Submitted 8 June, 2025; originally announced June 2025.

Semileptonic decay of double strangeness heavy flavor baryons

Authors: Hui-Hui Duan, Yong-Lu Liu, Qin Chang, Ming-Qiu Huang

Abstract: This paper investigates the double strangeness heavy flavor baryons $Ω_c^0$ and $Ω_b^-$, which contain two strange quarks. Using QCD light-cone sum rules (LCSRs), we calculate the form factors for the Cabibbo-suppressed processes $Ω_c^0\toΞ^-$ and $Ω_b^-\toΞ^0$, corresponding to the heavy-quark transitions $c\to d$ and $b\to u$, respectively. Combining these with the helicity amplitude formalism f… ▽ More This paper investigates the double strangeness heavy flavor baryons $Ω_c^0$ and $Ω_b^-$, which contain two strange quarks. Using QCD light-cone sum rules (LCSRs), we calculate the form factors for the Cabibbo-suppressed processes $Ω_c^0\toΞ^-$ and $Ω_b^-\toΞ^0$, corresponding to the heavy-quark transitions $c\to d$ and $b\to u$, respectively. Combining these with the helicity amplitude formalism for semileptonic decay differential widths, we computed the branching fractions of their corresponding semileptonic decay processes. Our analysis reveals significant discrepancies between two versions of QCD LCSRs: one using the light-cone distribution amplitudes (LCDAs) of the final-state $Ξ$ baryon and the other using the LCDAs of the initial-state double strangeness heavy flavor baryons. The results obtained with the $Ξ$ baryon's LCDAs show excellent agreement with other theoretical calculations. However, when using the LCDAs of the double strangeness heavy flavor baryons, the results differ by orders of magnitude, warranting further investigation. △ Less

Submitted 30 September, 2025; v1 submitted 27 May, 2025; originally announced May 2025.

Comments: 14 pages,6 figures

Journal ref: Phys.Rev.D 112 (2025) 5, 054030

Form Factors for $B_c^*\to η_c+lν_l$ at NLO in QCD

Authors: Wei Tao, Ya-Hui Zhao, Li-Ting Wang, Qin Chang, Zhen-Jun Xiao

Abstract: We present the Non-Relativistic QCD (NRQCD) calculations at the next-to-leading order (NLO) of $α_s$ for $B_c^*\to η_c$ vector, axial-vector, tensor and axial-tensor form factors, and obtain complete analytical expressions for the form factors, along with their asymptotic forms in the hierarchical heavy quark limit. Our results show that the NLO corrections are both sizable and well-behaved in the… ▽ More We present the Non-Relativistic QCD (NRQCD) calculations at the next-to-leading order (NLO) of $α_s$ for $B_c^*\to η_c$ vector, axial-vector, tensor and axial-tensor form factors, and obtain complete analytical expressions for the form factors, along with their asymptotic forms in the hierarchical heavy quark limit. Our results show that the NLO corrections are both sizable and well-behaved in the low squared transfer momentum $(q^2)$ region. Using the NRQCD + lattice + $z$-series method, we further provide theoretical predictions for $B_c^*\to η_c$ form factors across the full physical $q^2$ range. Based on these predicted form factors, we finally compute the decay widths and branching fractions for the semileptonic decays $B_c^*\to η_c+l{ν_l}$. △ Less

Submitted 17 March, 2026; v1 submitted 26 May, 2025; originally announced May 2025.

Comments: to be published in PRD

A Character-based Diffusion Embedding Algorithm for Enhancing the Generation Quality of Generative Linguistic Steganographic Texts

Authors: Yingquan Chen, Qianmu Li, Xiaocong Wu, Huifeng Li, Qing Chang

Abstract: Generating high-quality steganographic text is a fundamental challenge in the field of generative linguistic steganography. This challenge arises primarily from two aspects: firstly, the capabilities of existing models in text generation are limited; secondly, embedding algorithms fail to effectively mitigate the negative impacts of sensitive information's properties, such as semantic content or r… ▽ More Generating high-quality steganographic text is a fundamental challenge in the field of generative linguistic steganography. This challenge arises primarily from two aspects: firstly, the capabilities of existing models in text generation are limited; secondly, embedding algorithms fail to effectively mitigate the negative impacts of sensitive information's properties, such as semantic content or randomness. Specifically, to ensure that the recipient can accurately extract hidden information, embedding algorithms often have to consider selecting candidate words with relatively low probabilities. This phenomenon leads to a decrease in the number of high-probability candidate words and an increase in low-probability candidate words, thereby compromising the semantic coherence and logical fluency of the steganographic text and diminishing the overall quality of the generated steganographic material. To address this issue, this paper proposes a novel embedding algorithm, character-based diffusion embedding algorithm (CDEA). Unlike existing embedding algorithms that strive to eliminate the impact of sensitive information's properties on the generation process, CDEA leverages sensitive information's properties. It enhances the selection frequency of high-probability candidate words in the candidate pool based on general statistical properties at the character level and grouping methods based on power-law distributions, while reducing the selection frequency of low-probability candidate words in the candidate pool. Furthermore, to ensure the effective transformation of sensitive information in long sequences, we also introduce the XLNet model. Experimental results demonstrate that the combination of CDEA and XLNet significantly improves the quality of generated steganographic text, particularly in terms of perceptual-imperceptibility. △ Less

Submitted 7 May, 2025; v1 submitted 1 May, 2025; originally announced May 2025.

Comments: we need to clarify authorship and make further revisions in collaboration with co-authors

Enhancing Flexibility in Smart Manufacturing: System Property Enabled Multiagent Approach for Mobile Robot Scheduling in Multiproduct Flexible Manufacturing Systems

Authors: Muhammad Waseem, Qing Chang

Abstract: The present volatile global market is marked by a rising need for extensively tailored products. Flourishing in such an environment necessitates fostering a more intimate connection between market requirements and the manufacturing system, with a primary emphasis on the customer. Multiproduct flexible manufacturing systems (FMS) appear to be one of the most suitable manufacturing systems for mass-… ▽ More The present volatile global market is marked by a rising need for extensively tailored products. Flourishing in such an environment necessitates fostering a more intimate connection between market requirements and the manufacturing system, with a primary emphasis on the customer. Multiproduct flexible manufacturing systems (FMS) appear to be one of the most suitable manufacturing systems for mass-customization, given their capacity to readily adapt to variations in product specifications and functionalities to meet evolving market demands. Mobile robots play a pivotal role in managing multiproduct FMS. However, scheduling these robots poses a challenging problem due to system internal uncertainties and fluctuations in external demand. This paper proposes a solution to the robots' scheduling problem in a multiproduct FMS with stochastic machine failures and uncertain market demand for different product types. The proposed approach abstracts system properties for FMS and leverages these properties to improve the multiagent deep deterministic policy gradient (MADDPG) algorithm. This modification involves integrating parallel layers into the critic network, focusing specifically on crucial information such as varying customer demands. To validate the effectiveness of the proposed method, a comprehensive comparative analysis is conducted, comparing it with the simple-MADDPG, DDPG, and DQN algorithms. The results of the analysis demonstrate an average improvement of 9% in training time and 19% in market demand satisfaction. Moreover, to ensure scalability, transfer learning is employed, showcasing the applicability of the proposed method in large-scale complex environments. △ Less

Submitted 10 April, 2025; originally announced April 2025.

PRM-BAS: Enhancing Multimodal Reasoning through PRM-guided Beam Annealing Search

Authors: Pengfei Hu, Zhenrong Zhang, Qikai Chang, Shuhang Liu, Jiefeng Ma, Jun Du, Jianshu Zhang, Quan Liu, Jianqing Gao, Feng Ma, Qingfeng Liu

Abstract: Recent work increasingly focuses on improving the reasoning capabilities of Multimodal Large Language Models (MLLMs). Among existing methods, Process Reward Models (PRMs) stand out for offering dense, step-wise supervision to guide intermediate reasoning. However, how to effectively integrate PRMs into search strategies remains an open question. In this paper, we introduce PRM-BAS (PRM-Guided Beam… ▽ More Recent work increasingly focuses on improving the reasoning capabilities of Multimodal Large Language Models (MLLMs). Among existing methods, Process Reward Models (PRMs) stand out for offering dense, step-wise supervision to guide intermediate reasoning. However, how to effectively integrate PRMs into search strategies remains an open question. In this paper, we introduce PRM-BAS (PRM-Guided Beam Annealing Search), a lightweight approach for PRM-guided reasoning that dynamically adjusts beam size -- starting with a broader search space and gradually narrowing it as contextual information accumulates, thereby balancing performance and efficiency. We further propose a unified framework for data construction and PRM training. Specifically, we construct the PRM-BAS-300k dataset by selecting 300k questions from existing datasets and performing rollouts at each step to estimate the probability of reaching a correct final answer. The PRM is then trained using a combination of value loss for absolute action quality and rank loss for relative action quality. Extensive experiments on challenging multimodal reasoning benchmarks demonstrate that PRM-BAS significantly improves reasoning performance while maintaining low computational cost. Moreover, it generalizes well across different model scales and architectures, showcasing strong robustness and plug-and-play capability. △ Less

Submitted 14 April, 2025; originally announced April 2025.

EmoDiffusion: Enhancing Emotional 3D Facial Animation with Latent Diffusion Models

Authors: Yixuan Zhang, Qing Chang, Yuxi Wang, Guang Chen, Zhaoxiang Zhang, Junran Peng

Abstract: Speech-driven 3D facial animation seeks to produce lifelike facial expressions that are synchronized with the speech content and its emotional nuances, finding applications in various multimedia fields. However, previous methods often overlook emotional facial expressions or fail to disentangle them effectively from the speech content. To address these challenges, we present EmoDiffusion, a novel… ▽ More Speech-driven 3D facial animation seeks to produce lifelike facial expressions that are synchronized with the speech content and its emotional nuances, finding applications in various multimedia fields. However, previous methods often overlook emotional facial expressions or fail to disentangle them effectively from the speech content. To address these challenges, we present EmoDiffusion, a novel approach that disentangles different emotions in speech to generate rich 3D emotional facial expressions. Specifically, our method employs two Variational Autoencoders (VAEs) to separately generate the upper face region and mouth region, thereby learning a more refined representation of the facial sequence. Unlike traditional methods that use diffusion models to connect facial expression sequences with audio inputs, we perform the diffusion process in the latent space. Furthermore, we introduce an Emotion Adapter to evaluate upper face movements accurately. Given the paucity of 3D emotional talking face data in the animation industry, we capture facial expressions under the guidance of animation experts using LiveLinkFace on an iPhone. This effort results in the creation of an innovative 3D blendshape emotional talking face dataset (3D-BEF) used to train our network. Extensive experiments and perceptual evaluations validate the effectiveness of our approach, confirming its superiority in generating realistic and emotionally rich facial animations. △ Less

Submitted 13 March, 2025; originally announced March 2025.

Task-Specific Knowledge Distillation from the Vision Foundation Model for Enhanced Medical Image Segmentation

Authors: Pengchen Liang, Haishan Huang, Bin Pu, Jianguo Chen, Xiang Hua, Jing Zhang, Weibo Ma, Zhuangzhuang Chen, Yiwei Li, Qing Chang

Abstract: Large-scale pre-trained models, such as Vision Foundation Models (VFMs), have demonstrated impressive performance across various downstream tasks by transferring generalized knowledge, especially when target data is limited. However, their high computational cost and the domain gap between natural and medical images limit their practical application in medical segmentation tasks. Motivated by this… ▽ More Large-scale pre-trained models, such as Vision Foundation Models (VFMs), have demonstrated impressive performance across various downstream tasks by transferring generalized knowledge, especially when target data is limited. However, their high computational cost and the domain gap between natural and medical images limit their practical application in medical segmentation tasks. Motivated by this, we pose the following important question: "How can we effectively utilize the knowledge of large pre-trained VFMs to train a small, task-specific model for medical image segmentation when training data is limited?" To address this problem, we propose a novel and generalizable task-specific knowledge distillation framework. Our method fine-tunes the VFM on the target segmentation task to capture task-specific features before distilling the knowledge to smaller models, leveraging Low-Rank Adaptation (LoRA) to reduce the computational cost of fine-tuning. Additionally, we incorporate synthetic data generated by diffusion models to augment the transfer set, enhancing model performance in data-limited scenarios. Experimental results across five medical image datasets demonstrate that our method consistently outperforms task-agnostic knowledge distillation and self-supervised pretraining approaches like MoCo v3 and Masked Autoencoders (MAE). For example, on the KidneyUS dataset, our method achieved a 28% higher Dice score than task-agnostic KD using 80 labeled samples for fine-tuning. On the CHAOS dataset, it achieved an 11% improvement over MAE with 100 labeled samples. These results underscore the potential of task-specific knowledge distillation to train accurate, efficient models for medical image segmentation in data-constrained settings. △ Less

Submitted 10 March, 2025; originally announced March 2025.

Comments: 29 pages, 10 figures, 16 tables

Spatial-Temporal Perception with Causal Inference for Naturalistic Driving Action Recognition

Authors: Qing Chang, Wei Dai, Zhihao Shuai, Limin Yu, Yutao Yue

Abstract: Naturalistic driving action recognition is essential for vehicle cabin monitoring systems. However, the complexity of real-world backgrounds presents significant challenges for this task, and previous approaches have struggled with practical implementation due to their limited ability to observe subtle behavioral differences and effectively learn inter-frame features from video. In this paper, we… ▽ More Naturalistic driving action recognition is essential for vehicle cabin monitoring systems. However, the complexity of real-world backgrounds presents significant challenges for this task, and previous approaches have struggled with practical implementation due to their limited ability to observe subtle behavioral differences and effectively learn inter-frame features from video. In this paper, we propose a novel Spatial-Temporal Perception (STP) architecture that emphasizes both temporal information and spatial relationships between key objects, incorporating a causal decoder to perform behavior recognition and temporal action localization. Without requiring multimodal input, STP directly extracts temporal and spatial distance features from RGB video clips. Subsequently, these dual features are jointly encoded by maximizing the expected likelihood across all possible permutations of the factorization order. By integrating temporal and spatial features at different scales, STP can perceive subtle behavioral changes in challenging scenarios. Additionally, we introduce a causal-aware module to explore relationships between video frame features, significantly enhancing detection efficiency and performance. We validate the effectiveness of our approach using two publicly available driver distraction detection benchmarks. The results demonstrate that our framework achieves state-of-the-art performance. △ Less

Submitted 5 March, 2025; originally announced March 2025.

Pretrained LLMs as Real-Time Controllers for Robot Operated Serial Production Line

Authors: Muhammad Waseem, Kshitij Bhatta, Chen Li, Qing Chang

Abstract: The manufacturing industry is undergoing a transformative shift, driven by cutting-edge technologies like 5G, AI, and cloud computing. Despite these advancements, effective system control, which is crucial for optimizing production efficiency, remains a complex challenge due to the intricate, knowledge-dependent nature of manufacturing processes and the reliance on domain-specific expertise. Conve… ▽ More The manufacturing industry is undergoing a transformative shift, driven by cutting-edge technologies like 5G, AI, and cloud computing. Despite these advancements, effective system control, which is crucial for optimizing production efficiency, remains a complex challenge due to the intricate, knowledge-dependent nature of manufacturing processes and the reliance on domain-specific expertise. Conventional control methods often demand heavy customization, considerable computational resources, and lack transparency in decision-making. In this work, we investigate the feasibility of using Large Language Models (LLMs), particularly GPT-4, as a straightforward, adaptable solution for controlling manufacturing systems, specifically, mobile robot scheduling. We introduce an LLM-based control framework to assign mobile robots to different machines in robot assisted serial production lines, evaluating its performance in terms of system throughput. Our proposed framework outperforms traditional scheduling approaches such as First-Come-First-Served (FCFS), Shortest Processing Time (SPT), and Longest Processing Time (LPT). While it achieves performance that is on par with state-of-the-art methods like Multi-Agent Reinforcement Learning (MARL), it offers a distinct advantage by delivering comparable throughput without the need for extensive retraining. These results suggest that the proposed LLM-based solution is well-suited for scenarios where technical expertise, computational resources, and financial investment are limited, while decision transparency and system scalability are critical concerns. △ Less

Submitted 5 March, 2025; originally announced March 2025.

Comments: 20 pages, 7 figures

Rapid Bone Scintigraphy Enhancement via Semantic Prior Distillation from Segment Anything Model

Authors: Pengchen Liang, Leijun Shi, Huiping Yao, Bin Pu, Jianguo Chen, Lei Zhao, Haishan Huang, Zhuangzhuang Chen, Zhaozhao Xu, Lite Xu, Qing Chang, Yiwei Li

Abstract: Rapid bone scintigraphy is crucial for diagnosing skeletal disorders and detecting tumor metastases in children, as it shortens scan duration and reduces discomfort. However, accelerated acquisition often degrades image quality, impairing the visibility of fine anatomical details and potentially compromising diagnosis. To overcome this limitation, we introduce the first application of SAM-based se… ▽ More Rapid bone scintigraphy is crucial for diagnosing skeletal disorders and detecting tumor metastases in children, as it shortens scan duration and reduces discomfort. However, accelerated acquisition often degrades image quality, impairing the visibility of fine anatomical details and potentially compromising diagnosis. To overcome this limitation, we introduce the first application of SAM-based semantic priors for medical image restoration, utilizing the Segment Anything Model (SAM) to enhance pediatric rapid bone scintigraphy. Our approach employs two cascaded networks, $f^{IR1}$ and $f^{IR2}$, supported by three specialized modules: a Semantic Prior Integration (SPI) module, a Semantic Knowledge Distillation (SKD) module, and a Semantic Consistency Module (SCM). The SPI and SKD modules inject domain-specific semantic cues from a fine-tuned SAM, while the SCM preserves coherent semantic feature representations across both cascaded stages. Moreover, we present RBS, a novel Rapid Bone Scintigraphy dataset comprising paired standard (20 cm/min) and rapid (40 cm/min) scans from 137 pediatric patients aged 0.5 - 16 years, making it the first dataset tailored for pediatric rapid bone scintigraphy restoration. Extensive experiments on both a public endoscopic dataset and our RBS dataset demonstrate that our method consistently surpasses existing techniques in PSNR, SSIM, FID, and LPIPS metrics. △ Less

Submitted 4 June, 2025; v1 submitted 4 March, 2025; originally announced March 2025.

Comments: 12 pages, 9 figures, 8 tables

Next-to-leading order QCD corrections to $B_c^*\to J/ψ$ form factors

Authors: Qin Chang, Wei Tao, Zhen-Jun Xiao, Ruilin Zhu

Abstract: Within the framework of Non-Relativistic Quantum Chromodynamics (NRQCD) factorization, we calculate the next-to-leading order (NLO) perturbative QCD corrections to the form factors for the semileptonic decays of $B_c^*$ into $J/ψ$ via (axial-)vector and (axial-)tensor currents. We obtain the complete analytical results for the form factors up to NLO, and provide their asymptotic expressions in the… ▽ More Within the framework of Non-Relativistic Quantum Chromodynamics (NRQCD) factorization, we calculate the next-to-leading order (NLO) perturbative QCD corrections to the form factors for the semileptonic decays of $B_c^*$ into $J/ψ$ via (axial-)vector and (axial-)tensor currents. We obtain the complete analytical results for the form factors up to NLO, and provide their asymptotic expressions in the hierarchical heavy quark limit. The NLO corrections are found to be both significant and convergent in the relatively small squared transfer momentum ($q^2$) region, while also reducing the dependence on the renormalization scale $μ$. Finally, the theoretical predictions for $B_c^*\to J/ψ$ form factors over the full $q^2$ range are provided. △ Less

Submitted 27 February, 2025; originally announced February 2025.

Hybrid Robot Learning for Automatic Robot Motion Planning in Manufacturing

Authors: Siddharth Singh, Tian Yu, Qing Chang, John Karigiannis, Shaopeng Liu

Abstract: Industrial robots are widely used in diverse manufacturing environments. Nonetheless, how to enable robots to automatically plan trajectories for changing tasks presents a considerable challenge. Further complexities arise when robots operate within work cells alongside machines, humans, or other robots. This paper introduces a multi-level hybrid robot motion planning method combining a task space… ▽ More Industrial robots are widely used in diverse manufacturing environments. Nonetheless, how to enable robots to automatically plan trajectories for changing tasks presents a considerable challenge. Further complexities arise when robots operate within work cells alongside machines, humans, or other robots. This paper introduces a multi-level hybrid robot motion planning method combining a task space Reinforcement Learning-based Learning from Demonstration (RL-LfD) agent and a joint-space based Deep Reinforcement Learning (DRL) based agent. A higher level agent learns to switch between the two agents to enable feasible and smooth motion. The feasibility is computed by incorporating reachability, joint limits, manipulability, and collision risks of the robot in the given environment. Therefore, the derived hybrid motion planning policy generates a feasible trajectory that adheres to task constraints. The effectiveness of the method is validated through sim ulated robotic scenarios and in a real-world setup. △ Less

Submitted 26 February, 2025; originally announced February 2025.

Comments: 35 Pages, 11 Figures

Vision Foundation Models in Medical Image Analysis: Advances and Challenges

Authors: Pengchen Liang, Bin Pu, Haishan Huang, Yiwei Li, Hualiang Wang, Weibo Ma, Qing Chang

Abstract: The rapid development of Vision Foundation Models (VFMs), particularly Vision Transformers (ViT) and Segment Anything Model (SAM), has sparked significant advances in the field of medical image analysis. These models have demonstrated exceptional capabilities in capturing long-range dependencies and achieving high generalization in segmentation tasks. However, adapting these large models to medica… ▽ More The rapid development of Vision Foundation Models (VFMs), particularly Vision Transformers (ViT) and Segment Anything Model (SAM), has sparked significant advances in the field of medical image analysis. These models have demonstrated exceptional capabilities in capturing long-range dependencies and achieving high generalization in segmentation tasks. However, adapting these large models to medical image analysis presents several challenges, including domain differences between medical and natural images, the need for efficient model adaptation strategies, and the limitations of small-scale medical datasets. This paper reviews the state-of-the-art research on the adaptation of VFMs to medical image segmentation, focusing on the challenges of domain adaptation, model compression, and federated learning. We discuss the latest developments in adapter-based improvements, knowledge distillation techniques, and multi-scale contextual feature modeling, and propose future directions to overcome these bottlenecks. Our analysis highlights the potential of VFMs, along with emerging methodologies such as federated learning and model compression, to revolutionize medical image analysis and enhance clinical applications. The goal of this work is to provide a comprehensive overview of current approaches and suggest key areas for future research that can drive the next wave of innovation in medical image segmentation. △ Less

Submitted 20 February, 2025; v1 submitted 20 February, 2025; originally announced February 2025.

Comments: 17 pages, 1 figure

Topology-Aware Wavelet Mamba for Airway Structure Segmentation in Postoperative Recurrent Nasopharyngeal Carcinoma CT Scans

Authors: Haishan Huang, Pengchen Liang, Naier Lin, Luxi Wang, Bin Pu, Jianguo Chen, Qing Chang, Xia Shen, Guo Ran

Abstract: Nasopharyngeal carcinoma (NPC) patients often undergo radiotherapy and chemotherapy, which can lead to postoperative complications such as limited mouth opening and joint stiffness, particularly in recurrent cases that require re-surgery. These complications can affect airway function, making accurate postoperative airway risk assessment essential for managing patient care. Accurate segmentation o… ▽ More Nasopharyngeal carcinoma (NPC) patients often undergo radiotherapy and chemotherapy, which can lead to postoperative complications such as limited mouth opening and joint stiffness, particularly in recurrent cases that require re-surgery. These complications can affect airway function, making accurate postoperative airway risk assessment essential for managing patient care. Accurate segmentation of airway-related structures in postoperative CT scans is crucial for assessing these risks. This study introduces TopoWMamba (Topology-aware Wavelet Mamba), a novel segmentation model specifically designed to address the challenges of postoperative airway risk evaluation in recurrent NPC patients. TopoWMamba combines wavelet-based multi-scale feature extraction, state-space sequence modeling, and topology-aware modules to segment airway-related structures in CT scans robustly. By leveraging the Wavelet-based Mamba Block (WMB) for hierarchical frequency decomposition and the Snake Conv VSS (SCVSS) module to preserve anatomical continuity, TopoWMamba effectively captures both fine-grained boundaries and global structural context, crucial for accurate segmentation in complex postoperative scenarios. Through extensive testing on the NPCSegCT dataset, TopoWMamba achieves an average Dice score of 88.02%, outperforming existing models such as UNet, Attention UNet, and SwinUNet. Additionally, TopoWMamba is tested on the SegRap 2023 Challenge dataset, where it shows a significant improvement in trachea segmentation with a Dice score of 95.26%. The proposed model provides a strong foundation for automated segmentation, enabling more accurate postoperative airway risk evaluation. △ Less

Submitted 20 February, 2025; originally announced February 2025.

Comments: 20 pages, 11 figures, 6 tables

Interpretation of 95 GeV Excess within the Georgi-Machacek Model in Light of Positive Definiteness Constraints

Authors: Xiaokang Du, Huiling Liu, Qin Chang

Abstract: The recent observation of a di-photon excess around 95 GeV by the CMS and ATLAS Collaborations, along with the $b\bar{b}$ excess reported by the LEP Collaboration in the same mass region, has drawn significant interest in the possibility of new physics beyond the Standard Model (SM). The Georgi-Machacek (GM) model, which extends the Higgs sector of the SM by introducing additional triplet scalars… ▽ More The recent observation of a di-photon excess around 95 GeV by the CMS and ATLAS Collaborations, along with the $b\bar{b}$ excess reported by the LEP Collaboration in the same mass region, has drawn significant interest in the possibility of new physics beyond the Standard Model (SM). The Georgi-Machacek (GM) model, which extends the Higgs sector of the SM by introducing additional triplet scalars while preserving custodial symmetry at tree level, provides a compelling framework to explain both excesses simultaneously via a light custodial singlet Higgs. In this work, we investigate whether the GM model can still accommodate these excesses when taking into account newly proposed vacuum stability constraints, particularly the positive definiteness conditions. Our numerical analysis not only confirms the existence of a viable parameter space capable of explaining the 95 GeV excesses, but also demonstrates that, compared to traditional tree-level constraints at the electroweak scale, the positive definiteness conditions further expand the allowed parameter space, thereby enhancing the viability of the GM model. Furthermore, we emphasize that future collider experiments will play a crucial role in testing this interpretation by refining Higgs coupling measurements and searching for additional Higgs bosons. △ Less

Submitted 31 July, 2025; v1 submitted 10 February, 2025; originally announced February 2025.

Comments: 35 pages, 3 figures, 1 table; version published in Phys. Rev. D

Journal ref: Phys.Rev.D 112 (2025) 1, 015019

MuonSLab: A plastic scintillator based detector for muon measurement in the deep ocean

Authors: Jiacheng Wu, Weilun Huang, Ruike Cao, Qichao Chang, Wang Ding, Jingtao Huang, Liang Li, Xinchen Li, Hualin Mei, Cen Mo, Hengbin Shao, Wei Tian, Xinliang Tian, Yichen Tian, Xin Xiang, Donglian Xu, Fuyudi Zhang, Wei Zhi, Yiwei Zhu

Abstract: Atmospheric muons are important probes for studying primary cosmic rays and extensive air showers. Additionally, they constitute a significant background for many underground and deep-sea neutrino experiments, such as TRopIcal DEep-sea Neutrino Telescope (TRIDENT). Understanding the muon flux at various depths in the deep sea is essential for validating TRIDENT simulations and guiding the developm… ▽ More Atmospheric muons are important probes for studying primary cosmic rays and extensive air showers. Additionally, they constitute a significant background for many underground and deep-sea neutrino experiments, such as TRopIcal DEep-sea Neutrino Telescope (TRIDENT). Understanding the muon flux at various depths in the deep sea is essential for validating TRIDENT simulations and guiding the development of optimized trigger strategies. This paper introduces a novel device based on plastic scintillalors and silicon photomultipliers (SiPMs) named MuonSLab, which is designed to measure muon flux in the deep sea and has the potential to be extended to other atmospheric muon property measurements. We discuss the design and instrumentation of MuonSLab and present results from several muon flux measurements, demonstrating its sensitivity to muon detection and its stability during operations across multiple locations. △ Less

Submitted 1 May, 2025; v1 submitted 29 January, 2025; originally announced January 2025.

Skeleton and Font Generation Network for Zero-shot Chinese Character Generation

Authors: Mobai Xue, Jun Du, Zhenrong Zhang, Jiefeng Ma, Qikai Chang, Pengfei Hu, Jianshu Zhang, Yu Hu

Abstract: Automatic font generation remains a challenging research issue, primarily due to the vast number of Chinese characters, each with unique and intricate structures. Our investigation of previous studies reveals inherent bias capable of causing structural changes in characters. Specifically, when generating a Chinese character similar to, but different from, those in the training samples, the bias is… ▽ More Automatic font generation remains a challenging research issue, primarily due to the vast number of Chinese characters, each with unique and intricate structures. Our investigation of previous studies reveals inherent bias capable of causing structural changes in characters. Specifically, when generating a Chinese character similar to, but different from, those in the training samples, the bias is prone to either correcting or ignoring these subtle variations. To address this concern, we propose a novel Skeleton and Font Generation Network (SFGN) to achieve a more robust Chinese character font generation. Our approach includes a skeleton builder and font generator. The skeleton builder synthesizes content features using low-resource text input, enabling our technique to realize font generation independently of content image inputs. Unlike previous font generation methods that treat font style as a global embedding, we introduce a font generator to align content and style features on the radical level, which is a brand-new perspective for font generation. Except for common characters, we also conduct experiments on misspelled characters, a substantial portion of which slightly differs from the common ones. Our approach visually demonstrates the efficacy of generated images and outperforms current state-of-the-art font generation methods. Moreover, we believe that misspelled character generation have significant pedagogical implications and verify such supposition through experiments. We used generated misspelled characters as data augmentation in Chinese character error correction tasks, simulating the scenario where students learn handwritten Chinese characters with the help of misspelled characters. The significantly improved performance of error correction tasks demonstrates the effectiveness of our proposed approach and the value of misspelled character generation. △ Less

Submitted 14 January, 2025; originally announced January 2025.

Comments: 36 pages, 10 figures