arXiv subjects

Shuai Liu

Publications and source records attributed to Shuai Liu.

At least 73 records · Page 4Linked to original sources

arXiv2025

On the Formation of GW231123 in Population III Star Clusters

GW231123 is a binary black hole merger whose primary component lies within or above the pair-instability mass gap, while the secondary component falls within this gap. The standard theory of stellar evolution is significantly challenged by this event. We investigate the formation of candidate progenitors of GW231123 in Population III (Pop III) star clusters. We find that they could form through stellar mergers, binary black hole mergers, and mixed mergers. The mass distribution of these candidate progenitors covers the component masses of GW231123. Under our model assumptions, their predicted merger rate density spans the range of $0.001-0.26{\rm Gpc^{-3}yr^{-1}}$, encompassing that of GW231123. These findings suggest that GW231123 may originate from Pop III star clusters. Furthermore, such candidate progenitors are expected to be detectable by future gravitational wave detectors LISA/Taiji/TianQin/DECIGO/Cosmic Explorer/Einstein Telescope, which would provide valuable insights into the formation scenarios of events like GW231123.

astro-ph.GA↗

arXiv2025

High Performance Distributed Control for Large-Scale Linear Systems: A Cover-Based Distributed Observer Approach

In recent years, the distributed-observer-based distributed control law has shown powerful ability to arbitrarily approximate the centralized control performance. However, the traditional distributed observer requires each local observer to reconstruct the state information of the whole system, which is unrealistic for large-scale scenarios. To fill this gap, This paper presents a coverage solution algorithm for large-scale systems that accounts for both physical and communication network characteristics, which can significantly reduce the dimension of local observers. Then, the cover-based distributed observer for large-scale systems is proposed to overcome the problem that the system dynamics are difficult to estimate due to the coupling between cover sets. Furthermore, the two-layer Lyapunov analysis method is adopted and the dynamic transformation lemma of compact errors is proved, which solves the problem of analyzing stability of the error dynamic of the cover-based distributed observer. Finally, it is proved that the distributed control law based on the cover-based distributed observer can also arbitrarily approximate the control performance of the centralized control law, and the dimension of the local observer is greatly reduced compared with the traditional method. The simulation results show the validity of the developed theories.

eess.SY↗

arXiv2025

GaussianFusion: Gaussian-Based Multi-Sensor Fusion for End-to-End Autonomous Driving

Multi-sensor fusion is crucial for improving the performance and robustness of end-to-end autonomous driving systems. Existing methods predominantly adopt either attention-based flatten fusion or bird's eye view fusion through geometric transformations. However, these approaches often suffer from limited interpretability or dense computational overhead. In this paper, we introduce GaussianFusion, a Gaussian-based multi-sensor fusion framework for end-to-end autonomous driving. Our method employs intuitive and compact Gaussian representations as intermediate carriers to aggregate information from diverse sensors. Specifically, we initialize a set of 2D Gaussians uniformly across the driving scene, where each Gaussian is parameterized by physical attributes and equipped with explicit and implicit features. These Gaussians are progressively refined by integrating multi-modal features. The explicit features capture rich semantic and spatial information about the traffic scene, while the implicit features provide complementary cues beneficial for trajectory planning. To fully exploit rich spatial and semantic information in Gaussians, we design a cascade planning head that iteratively refines trajectory predictions through interactions with Gaussians. Extensive experiments on the NAVSIM and Bench2Drive benchmarks demonstrate the effectiveness and robustness of the proposed GaussianFusion framework. The source code will be released at https://github.com/Say2L/GaussianFusion.

cs.RO↗

arXiv2025

Evaluating the chromospheric structure model of AD Leo using RH1.5D and magnetic field data

Context. The interplay between surface magnetic topology and chromospheric heating in active M dwarfs remains poorly constrained, limiting our understanding of their magnetic cycles and high-energy environments. Aims. We aim to test whether detailed Zeeman-Doppler imaging (ZDI) maps of AD Leo can be used to spatially anchor a multi-component chromospheric model and validate the link between magnetic flux distribution and emission-line formation. Methods. We analyze high-resolution CARMENES spectra of H-alpha and the Ca II infrared triplet, together with ZDI maps. Synthetic profiles are computed using the RH1.5D non-LTE radiative transfer code with two active atmospheric components (low-latitude near the equator and polar near the pole) and a quiet background. Their relative filling factors and temperature structures are optimized per epoch. The ZDI maps serve as qualitative references for the large-scale magnetic topology but are not used as input to the optimization. Results. Our model reproduces the spectral line profiles across multiple epochs. The low-latitude active region shows notable variability, accounting for approximately 55-86% of the emission, while the polar region remains relatively constant in area (12-17%) but exhibits temperature variations over time, particularly during periods of increased activity. The spatial locations of the active regions derived from spectroscopy agree well with the radial magnetic field distribution from ZDI. Conclusions. Combining spectroscopic modeling with magnetic field maps is an effective approach for mapping magneto-chromospheric structures in M dwarfs. This framework deepens our understanding of stellar magnetic cycles and chromospheric dynamics, paving the way for detailed time-resolved studies in active low-mass stars.

astro-ph.SR↗

arXiv2025

Zero Echo Time Functional MRI in Humans

Motivation: Conventional echo planar imaging(EPI) based functional MRI(fMRI) uses the BOLD contrast to map activity changes in human brains. Introducing an efficient ZTE sequence for functional brain mapping can help address limitations of EPI and demonstrate the feasibility of using T1 related changes as a surrogate marker of brain activity. Goals: To test and optimize ZTE sequence for fMRI. Methods: A ZTE sequence with radial inside out spokes was used to prepare a dynamic imaging protocol that matches conventional EPI time course. Temporal SNR and sensitivity to susceptibility differences of ZTE were evaluated and the sequence was benchmarked against BOLD EPI in a task based visual fMRI study with healthy volunteers at 3T. Results: Phantom measurements confirmed sensitivity of the ZTE protocol to the oxygen concentration. Functional activation in primary visual cortex could be detected using ZTE. Resting state networks could also be identified using independent component analysis. Discussion: ZTE-based fMRI is proposed for mapping functional activation in human brain. ZTE is robust against susceptibility artefacts and significantly reduces acoustic noise. Radial sampling pattern allows for high undersampling rates to increase temporal resolution.

physics.med-ph↗

arXiv2025

Enhancing Efficiency and Exploration in Reinforcement Learning for LLMs

Reasoning large language models (LLMs) excel in complex tasks, which has drawn significant attention to reinforcement learning (RL) for LLMs. However, existing approaches allocate an equal number of rollouts to all questions during the RL process, which is inefficient. This inefficiency stems from the fact that training on simple questions yields limited gains, whereas more rollouts are needed for challenging questions to sample correct answers. Furthermore, while RL improves response precision, it limits the model's exploration ability, potentially resulting in a performance cap below that of the base model prior to RL. To address these issues, we propose a mechanism for dynamically allocating rollout budgets based on the difficulty of the problems, enabling more efficient RL training. Additionally, we introduce an adaptive dynamic temperature adjustment strategy to maintain the entropy at a stable level, thereby encouraging sufficient exploration. This enables LLMs to improve response precision while preserving their exploratory ability to uncover potential correct pathways. The code and data is available on: https://github.com/LiaoMengqi/E3-RL4LLMs

cs.LG↗

arXiv2025

A Survey of Defenses Against AI-Generated Visual Media: Detection,Disruption, and Authentication

Deep generative models have demonstrated impressive performance in various computer vision applications, including image synthesis, video generation, and medical analysis. Despite their significant advancements, these models may be used for malicious purposes, such as misinformation, deception, and copyright violation. In this paper, we provide a systematic and timely review of research efforts on defenses against AI-generated visual media, covering detection, disruption, and authentication. We review existing methods and summarize the mainstream defense-related tasks within a unified passive and proactive framework. Moreover, we survey the derivative tasks concerning the trustworthiness of defenses, such as their robustness and fairness. For each defense strategy, we formulate its general pipeline and propose a multidimensional taxonomy applicable across defense tasks, based on methodological strategies. Additionally, we summarize the commonly used evaluation datasets, criteria, and metrics. Finally, by analyzing the reviewed studies, we provide insights into current research challenges and suggest possible directions for future research.

cs.CV↗

arXiv2025

NextLocLLM: Location Semantics Modeling and Coordinate-Based Next Location Prediction with LLMs

Next location prediction is a critical task in human mobility analysis.Existing methods typically formulate it as a classification task based on discrete location IDs, which hinders spatial continuity modeling and limits generalization to new cities. In this paper, we propose NextLocLLM, a novel framework that reformulates next-location prediction as coordinate regression and integrates LLMs for both location semantics encoding and coordinate-level prediction. To model location functional semantics, it constructs LLM-enhanced POI embeddings by leveraging language understanding capabilities of LLMs to extract functional semantics from textual descriptions of POI categories. These POI embeddings are combined with spatiotemporal trajectory representation and fed into the same LLM, enabling unified semantic and predictive modeling. A lightweight regression head generates coordinate outputs, which are mapped to top-k candidate locations via post-prediction retrieval module, ensuring structured outputs. Experiments across diverse cities show that NextLocLLM outperforms existing baselines in both supervised and zero-shot settings. Code is available at: https://github.com/liuwj2000/NexelocLLM.

cs.LG↗

arXiv2025

Reduced-Order Model-Guided Reinforcement Learning for Demonstration-Free Humanoid Locomotion

We introduce Reduced-Order Model-Guided Reinforcement Learning (ROM-GRL), a two-stage reinforcement learning framework for humanoid walking that requires no motion capture data or elaborate reward shaping. In the first stage, a compact 4-DOF (four-degree-of-freedom) reduced-order model (ROM) is trained via Proximal Policy Optimization. This generates energy-efficient gait templates. In the second stage, those dynamically consistent trajectories guide a full-body policy trained with Soft Actor--Critic augmented by an adversarial discriminator, ensuring the student's five-dimensional gait feature distribution matches the ROM's demonstrations. Experiments at 1 meter-per-second and 4 meter-per-second show that ROM-GRL produces stable, symmetric gaits with substantially lower tracking error than a pure-reward baseline. By distilling lightweight ROM guidance into high-dimensional policies, ROM-GRL bridges the gap between reward-only and imitation-based locomotion methods, enabling versatile, naturalistic humanoid behaviors without any human demonstrations.

cs.RO↗

arXiv2025

STRATA-TS: Selective Knowledge Transfer for Urban Time Series Forecasting with Retrieval-Guided Reasoning

Urban forecasting models often face a severe data imbalance problem: only a few cities have dense, long-span records, while many others expose short or incomplete histories. Direct transfer from data-rich to data-scarce cities is unreliable because only a limited subset of source patterns truly benefits the target domain, whereas indiscriminate transfer risks introducing noise and negative transfer. We present STRATA-TS (Selective TRAnsfer via TArget-aware retrieval for Time Series), a framework that combines domain-adapted retrieval with reasoning-capable large models to improve forecasting in scarce data regimes. STRATA-TS employs a patch-based temporal encoder to identify source subsequences that are semantically and dynamically aligned with the target query. These retrieved exemplars are then injected into a retrieval-guided reasoning stage, where an LLM performs structured inference over target inputs and retrieved support. To enable efficient deployment, we distill the reasoning process into a compact open model via supervised fine-tuning. Extensive experiments on three parking availability datasets across Singapore, Nottingham, and Glasgow demonstrate that STRATA-TS consistently outperforms strong forecasting and transfer baselines, while providing interpretable knowledge transfer pathways.

cs.LG↗

arXiv2025

Distributed Stochastic Zeroth-Order Optimization with Compressed Communication

The dual challenges of prohibitive communication overhead and the impracticality of gradient computation due to data privacy or black-box constraints in distributed systems motivate this work on communication-constrained gradient-free optimization. We propose a stochastic distributed zeroth-order algorithm (Com-DSZO) requiring only two function evaluations per iteration, integrated with general compression operators. Rigorous analysis establishes its sublinear convergence rate for both smooth and nonsmooth objectives, while explicitly elucidating the compression-convergence trade-off. Furthermore, we develop a variance-reduced variant (VR-Com-DSZO) under stochastic mini-batch feedback. The empirical algorithm performance are illustrated with numerical examples.

math.OC↗

arXiv2025

Wafer-Scale Squeezed-Light Chips

Squeezed-light generation in photonic integrated circuits (PICs) is essential for scalable continuous-variable (CV) quantum information processing. By suppressing quantum fluctuations below the shot-noise limit, squeezed states enable quantum-enhanced sensing and serve as a standard resource for CV quantum information processing. While chip-level squeezed-light sources have been demonstrated, extending this capability to the wafer level with reproducible strong squeezing to bolster large-scale quantum-enhanced sensing and information processing has been hindered by squeezed light's extreme susceptibility to device imperfections. Here, we report wafer-scale fabrication, generation, and characterization of two-mode squeezed-vacuum states on a fully complementary metal-oxide-semiconductor (CMOS)-compatible silicon nitride (Si$_3$N$_4$) PIC platform. Across a 4-inch wafer, 8 dies yield 2.9-3.1 dB directly measured quadrature squeezing with $< 0.2$ dB variation, demonstrating excellent uniformity. This performance is enabled by co-integrating ultralow-loss, strongly overcoupled high-$Q$ microresonators, cascaded pump-rejection filters, and low-loss inverse-tapered edge couplers. The measurements agree with a first-principles theoretical model parameterized solely by independently extracted device parameters and experimental settings. The measured squeezing level can be further improved by enhancing the efficiencies of off-chip detection and chip-to-fiber coupling. These results establish a reproducible, wafer-scale route to nonclassical-light generation in integrated photonics and lay the groundwork for scalable CV processors, multiplexed entanglement sources, and quantum-enhanced sensing.

quant-ph↗

arXiv2025

AIM 2025 Challenge on High FPS Motion Deblurring: Methods and Results

This paper presents a comprehensive review of the AIM 2025 High FPS Non-Uniform Motion Deblurring Challenge, highlighting the proposed solutions and final results. The objective of this challenge is to identify effective networks capable of producing clearer and visually compelling images in diverse and challenging conditions, by learning representative visual cues for complex aggregations of motion types. A total of 68 participants registered for the competition, and 9 teams ultimately submitted valid entries. This paper thoroughly evaluates the state-of-the-art advances in high-FPS single image motion deblurring, showcasing the significant progress in the field, while leveraging samples of the novel dataset, MIORe, that introduces challenging examples of movement patterns.

cs.CV↗

arXiv2025

AI-Powered Reconstruction of Dark Matter Velocity Fields from Redshift-Space Halo Distribution

We propose a UNet-based deep learning model to reconstruct the real-space dark matter (DM) velocity field from the redshift-space distribution of sparse DM halos. Using various statistical measures, we show that the reconstructed velocity components--including velocity magnitude, momentum, and divergence--closely match the ground truth, achieving better than 10% relative error and a correlation coefficient of 0.88. In the power spectrum comparison over $k \in [0.05, 0.3] h/{\rm Mpc}$, the UNet reconstruction outperforms linear theory and agrees with the true field within $2σ$. The model also effectively corrects redshift-space distortions (RSD), yielding unbiased power spectrum multipoles of DM fields within $2σ$. Notably, the UNet remains robust even with incomplete halo mass information. These results highlight the model's broad applicability to cosmological analyses, including RSD, cosmic web studies, the kinetic Sunyaev-Zel'dovich effect, and BAO reconstruction.

astro-ph.CO↗

arXiv2025

Spectral Efficiency-Aware Codebook Design for Task-Oriented Semantic Communications

Digital task-oriented semantic communication (ToSC) aims to transmit only task-relevant information, significantly reducing communication overhead. Existing ToSC methods typically rely on learned codebooks to encode semantic features and map them to constellation symbols. However, these codebooks are often sparsely activated, resulting in low spectral efficiency and underutilization of channel capacity. This highlights a key challenge: how to design a codebook that not only supports task-specific inference but also approaches the theoretical limits of channel capacity. To address this challenge, we construct a spectral efficiency-aware codebook design framework that explicitly incorporates the codebook activation probability into the optimization process. Beyond maximizing task performance, we introduce the Wasserstein (WS) distance as a regularization metric to minimize the gap between the learned activation distribution and the optimal channel input distribution. Furthermore, we reinterpret WS theory from a generative perspective to align with the semantic nature of ToSC. Combining the above two aspects, we propose a WS-based adaptive hybrid distribution scheme, termed WS-DC, which learns compact, task-driven and channel-aware latent representations. Experimental results demonstrate that WS-DC not only outperforms existing approaches in inference accuracy but also significantly improves codebook efficiency, offering a promising direction toward capacity-approaching semantic communication systems.

eess.SP↗

arXiv2025

D3: Training-Free AI-Generated Video Detection Using Second-Order Features

The evolution of video generation techniques, such as Sora, has made it increasingly easy to produce high-fidelity AI-generated videos, raising public concern over the dissemination of synthetic content. However, existing detection methodologies remain limited by their insufficient exploration of temporal artifacts in synthetic videos. To bridge this gap, we establish a theoretical framework through second-order dynamical analysis under Newtonian mechanics, subsequently extending the Second-order Central Difference features tailored for temporal artifact detection. Building on this theoretical foundation, we reveal a fundamental divergence in second-order feature distributions between real and AI-generated videos. Concretely, we propose Detection by Difference of Differences (D3), a novel training-free detection method that leverages the above second-order temporal discrepancies. We validate the superiority of our D3 on 4 open-source datasets (Gen-Video, VideoPhy, EvalCrafter, VidProM), 40 subsets in total. For example, on GenVideo, D3 outperforms the previous best method by 10.39% (absolute) mean Average Precision. Additional experiments on time cost and post-processing operations demonstrate D3's exceptional computational efficiency and strong robust performance. Our code is available at https://github.com/Zig-HS/D3.

cs.CV↗

arXiv2025

Hybrid Scandium Aluminum Nitride/Silicon Nitride Integrated Photonic Circuits

Scandium-doped aluminum nitride has recently emerged as a promising material for quantum photonic integrated circuits (PICs) due to its unique combination of strong second-order nonlinearity, ferroelectricity, piezoelectricity, and complementary metal-oxide-semiconductor (CMOS) compatibility. However, the relatively high optical loss reported to date-typically above 2.4 dB/cm-remains a key challenge that limits its widespread application in low-loss PICs. Here, we present a monolithically integrated $\mathrm{Si}_3\mathrm{N}_4$-ScAlN waveguide platform that overcomes this limitation. By confining light within an etched $\mathrm{Si}_3\mathrm{N}_4$ waveguide while preserving the functional properties of the underlying ScAlN layer, we achieve an intrinsic quality factor of $Q_{\mathrm{i}} = 3.35 \times 10^5$, corresponding to a propagation loss of 1.03 dB/cm-comparable to that of commercial single-mode silicon-on-insulator (SOI) waveguides. This hybrid architecture enables low-loss and scalable fabrication while retaining the advanced functionalities offered by ScAlN, such as ferroelectricity and piezoelectricity. Our results establish a new pathway for ScAlN-based PICs with potential applications in high-speed optical communication, modulation, sensing, nonlinear optics, and quantum optics within CMOS-compatible platforms.

physics.optics↗

arXiv2025

CD-NGP: A Fast Scalable Continual Representation for Dynamic Scenes

Novel view synthesis (NVS) in dynamic scenes faces persistent challenges in memory consumption, model complexity, training efficiency, and rendering quality. Offline methods offer high fidelity but suffer from high memory usage and limited scalability, while online approaches often trade quality for speed and compactness. We propose Continual Dynamic Neural Graphics Primitives (CD-NGP), a continual learning framework that reduces memory overhead and enhances scalability through parameter reuse. To avoid feature interference in dynamic scenes and improve rendering quality, our method combines spatial and temporal hash encodings, which compactly represent scene structures and motion patterns. We also introduce a new dataset comprising multi-view, long-duration ($>1200$ frames) videos with both rigid and non-rigid motion, which is not found in existing benchmarks. CD-NGP is evaluated on public datasets and our long video dataset, demonstrating superior scalability and reconstruction quality. It significantly reduces training memory usage to <14GB and requires only 0.4MB/frame in streaming bandwidth on DyNeRF -- substantially lower than most online baselines.

cs.CV↗