Search

Operating atom-interferometer gyroscopes outside a laboratory environment is challenging primarily owing to the instability of laser systems. To enhance the thermal stability of free-space laser systems, a compact laser system using fiber lasers and all-quartz-jointed optical modules was developed for a dual-atom-interferometer gyroscope. Millimeter-scale optical elements jointed on quartz plates with identical quartz supports, ensure laser power stability and facilitate component upgrades. The primary diode laser was locked to the modulation transfer spectrum of Rb atoms, and Raman lasers were phase-locked to the primary laser. Frequencies for repumping, blow-away, and detection lasers were adjusted with acousto-optic modulators. At room temperature, laser power fluctuation was under 1:1000, polarization extinction ratio exceeded 30 dB, frequency fluctuation was below 91 kHz, and phase noise reached to -100 dBc/Hz @ 1 kHz. The optical modules were tested at 5--50 $^{\circ}$C and applied to a dual-atom-interferometer gyroscope. The fringe contrast was tested over the temperature range. The proposed system paves the way for promoting field applications of atom-interferometer sensors., Comment: 11 pages, 9 figures

Traditional neural networks (multi-layer perceptrons) have become an important tool in data science due to their success across a wide range of tasks. However, their performance is sometimes unsatisfactory, and they often require a large number of parameters, primarily due to their reliance on the linear combination structure. Meanwhile, additive regression has been a popular alternative to linear regression in statistics. In this work, we introduce novel deep neural networks that incorporate the idea of additive regression. Our neural networks share architectural similarities with Kolmogorov-Arnold networks but are based on simpler yet flexible activation and basis functions. Additionally, we introduce several hybrid neural networks that combine this architecture with that of traditional neural networks. We derive their universal approximation properties and demonstrate their effectiveness through simulation studies and a real-data application. The numerical results indicate that our neural networks generally achieve better performance than traditional neural networks while using fewer parameters., Comment: 29 pages, 13 figures

We present a verifier of quantum programs called AutoQ 2.0. Quantum programs extend quantum circuits (the domain of AutoQ 1.0) by classical control flow constructs, which enable users to describe advanced quantum algorithms in a formal and precise manner. The extension is highly non-trivial, as we needed to tackle both theoretical challenges (such as the treatment of measurement, the normalization problem, and lifting techniques for verification of classical programs with loops to the quantum world), and engineering issues (such as extending the input format with a~support for specifying loop invariants). We have successfully used AutoQ 2.0 to verify two types of advanced quantum programs that cannot be expressed using only quantum circuits: the \emph{repeat-until-success} (RUS) algorithm and the weak-measurement-based version of Grover's search algorithm. AutoQ 2.0 can efficiently verify all our benchmarks: all RUS algorithms were verified instantly and, for the weak-measurement-based version of Grover's search, we were able to handle the case of 100 qubits in $\sim$20 minutes., Comment: regular tool paper submitted to TACAS 2025

Context. Photochemistry is a key process driving planetary atmospheres away from local thermodynamic equilibrium. Recent observations of the H$_2$ dominated atmospheres of hot gas giants have detected SO$_2$ as one of the major products of this process. Aims. We investigate which chemical pathways lead to the formation of SO$_2$ in an atmosphere, and we investigate which part of the flux from the host star is necessary to initiate SO$_2$ production. Methods. We use the publicly available S-N-C-H-O photochemical network in the VULCAN chemical kinetics code to compute the disequilibrium chemistry of an exoplanetary atmosphere. Results. We find that there are two distinct chemical pathways that lead to the formation of SO$_2$. The formation of SO$_2$ at higher pressures is initiated by stellar flux >200 nm, whereas the formation of SO$_2$ at lower pressures is initiated by stellar flux <200 nm. In deeper layers of the atmosphere, OH is provided by the hydrogen abstraction of H$_2$O, and sulfur is provided by the photodissociation of SH and S$_2$, which leads to a positive feedback cycle that liberates sulfur from the stable H$_2$S molecule. In higher layers of the atmosphere, OH is provided by the photodissociation of H$_2$O, and sulfur can be liberated from H$_2$S by either photodissociation of SH and S$_2$, or by the hydrogen abstraction of SH. Conclusions. We conclude that the stellar flux in the 200-350 nm wavelength range as well as the ratio of NUV/UV radiation are important parameters determining the observability of SO$_2$. In addition we find that there is a diversity of chemical pathways to the formation of SO$_2$. This is crucial for the interpretation of SO$_2$ detections and derived elemental abundance ratios and overall metallicities., Comment: Accepted for publication in Astronomy & Astrophysics

We introduce an industrial Head Blending pipeline for the task of seamlessly integrating an actor's head onto a target body in digital content creation. The key challenge stems from discrepancies in head shape and hair structure, which lead to unnatural boundaries and blending artifacts. Existing methods treat foreground and background as a single task, resulting in suboptimal blending quality. To address this problem, we propose CHANGER, a novel pipeline that decouples background integration from foreground blending. By utilizing chroma keying for artifact-free background generation and introducing Head shape and long Hair augmentation ($H^2$ augmentation) to simulate a wide range of head shapes and hair styles, CHANGER improves generalization on innumerable various real-world cases. Furthermore, our Foreground Predictive Attention Transformer (FPAT) module enhances foreground blending by predicting and focusing on key head and body regions. Quantitative and qualitative evaluations on benchmark datasets demonstrate that our CHANGER outperforms state-of-the-art methods, delivering high-fidelity, industrial-grade results., Comment: Accepted by WACV 2025. Project page: https://hahminlew.github.io/changer

Despite continuous advancements in deep learning for understanding human motion, existing models often struggle to accurately identify action timing and specific body parts, typically supporting only single-round interaction. Such limitations in capturing fine-grained motion details reduce their effectiveness in motion understanding tasks. In this paper, we propose MoChat, a multimodal large language model capable of spatio-temporal grounding of human motion and understanding multi-turn dialogue context. To achieve these capabilities, we group the spatial information of each skeleton frame based on human anatomical structure and then apply them with Joints-Grouped Skeleton Encoder, whose outputs are combined with LLM embeddings to create spatio-aware and temporal-aware embeddings separately. Additionally, we develop a pipeline for extracting timestamps from skeleton sequences based on textual annotations, and construct multi-turn dialogues for spatially grounding. Finally, various task instructions are generated for jointly training. Experimental results demonstrate that MoChat achieves state-of-the-art performance across multiple metrics in motion understanding tasks, making it as the first model capable of fine-grained spatio-temporal grounding of human motion.

Improving the efficiency and performance of implicit neural representations in 3D, particularly Neural Radiance Fields (NeRF) and Signed Distance Fields (SDF) is crucial for enabling their use in real-time applications. These models, while capable of generating photo-realistic novel views and detailed 3D reconstructions, often suffer from high computational costs and slow inference times. To address this, we introduce a novel neural network layer called the "magnituder", designed to reduce the number of training parameters in these models without sacrificing their expressive power. By integrating magnituders into standard feed-forward layer stacks, we achieve improved inference speed and adaptability. Furthermore, our approach enables a zero-shot performance boost in trained implicit neural representation models through layer-wise knowledge transfer without backpropagation, leading to more efficient scene reconstruction in dynamic environments.

Scaling up quantum dots to two-dimensional (2D) arrays is a crucial step for advancing semiconductor quantum computation. However, maintaining excellent tunability of quantum dot parameters, including both nearest-neighbor and next-nearest-neighbor couplings, during 2D scaling is challenging, particularly for silicon quantum dots due to their relatively small size. Here, we present a highly controllable and interconnected 2D quantum dot array in planar silicon, demonstrating independent control over electron fillings and the tunnel couplings of nearest-neighbor dots. More importantly, we also demonstrate the wide tuning of tunnel couplings between next-nearest-neighbor dots,which plays a crucial role in 2D quantum dot arrays. This excellent tunability enables us to alter the coupling configuration of the array as needed. These results open up the possibility of utilizing silicon quantum dot arrays as versatile platforms for quantum computing and quantum simulation.

Electron spin qubits in silicon are a promising platform for fault-tolerant quantum computing. Low-frequency noise, including nuclear spin fluctuations and charge noise, is a primary factor limiting gate fidelities. Suppressing this noise is crucial for high-fidelity qubit operations. Here, we report on a two-qubit quantum device in natural silicon with universal qubit control, designed to investigate the upper limits of gate fidelities in a non-purified Si/SiGe quantum dot device. By employing advanced device structures, qubit manipulation techniques, and optimization methods, we have achieved single-qubit gate fidelities exceeding 99% and a two-qubit Controlled-Z (CZ) gate fidelity of 91%. Decoupled CZ gates are used to prepare Bell states with a fidelity of 91%, typically exceeding previously reported values in natural silicon devices. These results underscore that even natural silicon has the potential to achieve high-fidelity gate operations, particularly with further optimization methods to suppress low-frequency noise.

Electronic warfare support (ES) systems intercept adversary radar signals and estimate various types of signal information, including modulation schemes. The accurate and rapid identification of modulation schemes under conditions of very low signal power remains a significant challenge for ES systems. This paper proposes a recognition model based on a noise-aware ensemble learning (NAEL) framework to efficiently recognize radar modulation schemes in noisy environments. The NAEL framework evaluates the influence of noise on recognition and adaptively selects an appropriate neural network structure, offering significant advantages in terms of computational efficiency and recognition performance. Furthermore, we employ feature extraction blocks to enhance the efficiency of the proposed recognition model. We present the analysis results of the recognition performance of the proposed model based on experimental data. Our recognition model demonstrates superior recognition accuracy with low computational complexity compared to conventional classification models., Comment: 10 pages, 10 figures

The rising demand for remote-sensing systems for detecting hazardous situations has led to increased interest in radar-based human activity recognition (HAR). Conventional radar-based HAR methods predominantly rely on micro-Doppler spectrograms for recognition tasks. However, spectrograms frequently fail to effectively capture micro-Doppler signatures because of their limited linear resolution. To address this limitation, we propose a time--frequency domain representation method that adaptively adjusts the resolution based on activity characteristics. This approach nonlinearly transforms the resolution to focus on the most relevant frequency range for micro-Doppler signatures. We validate the proposed method by training deep-learning-based HAR models on datasets generated using the adaptive representation method. Experimental results demonstrate that the models trained using the proposed method achieve superior recognition accuracy compared with those trained using conventional methods., Comment: 5 pages, 5 figures

We present the application of the image coaddition algorithm, Up-sampling and PSF Deconvolution Coaddition (UPDC), for stacking multiple exposure images captured by the James Webb Space Telescope (JWST) Near-Infrared Camera (NIRCam). By addressing the point spread function (PSF) effect, UPDC provides visually enhanced and sharper images. Furthermore, the anti-aliasing and super-resolution capabilities of UPDC make it easier to deblend sources overlapped on images, yielding a higher accuracy of aperture photometry. We apply this algorithm to the SMACS J0723 imaging data. Comparative analysis with the Drizzle algorithm demonstrates significant improvements in detecting faint sources, achieving accurate photometry, and effectively deblending (super-resolution) closely packed sources. {As a result, we have newly detected a pair of close binary stars that were previously unresolvable in the original exposures or the Drizzled image.} These improvements significantly benefit various scientific projects conducted by JWST. The resulting dataset, named "UPdec-Webb", can be accessible through the official website of the Chinese Virtual Observatory (ChinaVO)., Comment: 21 pages, 18 figures, accepted for publication in ApJS

Real-time visual feedback is essential for tetherless control of remotely operated vehicles, particularly during inspection and manipulation tasks. Though acoustic communication is the preferred choice for medium-range communication underwater, its limited bandwidth renders it impractical to transmit images or videos in real-time. To address this, we propose a model-based image compression technique that leverages prior mission information. Our approach employs trained machine-learning based novel view synthesis models, and uses gradient descent optimization to refine latent representations to help generate compressible differences between camera images and rendered images. We evaluate the proposed compression technique using a dataset from an artificial ocean basin, demonstrating superior compression ratios and image quality over existing techniques. Moreover, our method exhibits robustness to introduction of new objects within the scene, highlighting its potential for advancing tetherless remotely operated vehicle operations., Comment: Preprint submitted to Ocean Engineering

While 3D Gaussian Splatting (3DGS) has recently demonstrated remarkable rendering quality and efficiency in 3D scene reconstruction, it struggles with varying lighting conditions and incidental occlusions in real-world scenarios. To accommodate varying lighting conditions, existing 3DGS extensions apply color mapping to the massive Gaussian primitives with individually optimized appearance embeddings. To handle occlusions, they predict pixel-wise uncertainties via 2D image features for occlusion capture. Nevertheless, such massive color mapping and pixel-wise uncertainty prediction strategies suffer from not only additional computational costs but also coarse-grained lighting and occlusion handling. In this work, we propose a nexus kernel-driven approach, termed NexusSplats, for efficient and finer 3D scene reconstruction under complex lighting and occlusion conditions. In particular, NexusSplats leverages a novel light decoupling strategy where appearance embeddings are optimized based on nexus kernels instead of massive Gaussian primitives, thus accelerating reconstruction speeds while ensuring local color consistency for finer textures. Additionally, a Gaussian-wise uncertainty mechanism is developed, aligning 3D structures with 2D image features for fine-grained occlusion handling. Experimental results demonstrate that NexusSplats achieves state-of-the-art rendering quality while reducing reconstruction time by up to 70.4% compared to the current best in quality., Comment: submitted to CVPR 2025

Multimodal Large Language Models (MLLMs) are widely used for visual perception, understanding, and reasoning. However, long video processing and precise moment retrieval remain challenging due to LLMs' limited context size and coarse frame extraction. We propose the Large Language-and-Vision Assistant for Moment Retrieval (LLaVA-MR), which enables accurate moment retrieval and contextual grounding in videos using MLLMs. LLaVA-MR combines Dense Frame and Time Encoding (DFTE) for spatial-temporal feature extraction, Informative Frame Selection (IFS) for capturing brief visual and motion patterns, and Dynamic Token Compression (DTC) to manage LLM context limitations. Evaluations on benchmarks like Charades-STA and QVHighlights demonstrate that LLaVA-MR outperforms 11 state-of-the-art methods, achieving an improvement of 1.82% in R1@0.5 and 1.29% in mAP@0.5 on the QVHighlights dataset. Our implementation will be open-sourced upon acceptance.

Magnetic kagome metals have attracted tremendous research interests recently, because they represent an ideal playground for exploring the fascinating interplay between their intrinsically inherited topologically non-trivial electron band structures, magnetism and electronic correlation effects, and the resultant novel electronic/magnetic states and emergent excitations. In this work, we report a comprehensive single-crystal neutron diffraction investigation of the ground-state magnetic structures of the recently discovered V-based topological kagome metals RV$_6$Sn$_6$ (R = Tb, Dy, Ho, Er). Furthermore, the sample synthesis details and our systematic studies of crystal structure, low-temperature magnetic and thermodynamic properties of these compounds via various in-house characterization techniques are also reported. It can be revealed that RV$_6$Sn$_6$ (R = Tb, Dy, Ho) have a collinear ferromagnetic order in the ground state, with the ordered magnetic moment aligned along the c axis for R = Tb, Ho, while approximately 20${^\circ}$ tilted off from the c axis for R = Dy. In contrast, ErV$_6$Sn$_6$ shows an A-type antiferromagnetic structure with a magnetic propagation vector k = (0, 0, 0.5), and with the ordered magnetic moment aligned in the ab plane. A comparison of the low-temperature magnetic structures for both the extensively investigated topological kagome metal series of RV$_6$Sn$_6$ and RMn$_6$Sn$_6$ is given in details. This allows to gain new insights into the complex magnetic interactions, diverse single-ion magnetic anisotropies and spin dynamics in these compounds. The reported ground-state magnetic structures in RV$_6$Sn$_6$ (R = Tb, Dy, Ho, Er) can pave the way for further explorations of the possible interplay between magnetism and topologically non-trivial electron band structures in the magnetically ordered phase regime., Comment: To appear in Phys. Rev. Research

While large language models (LLMs) have achieved notable success in generative tasks, they still face limitations, such as lacking up-to-date knowledge and producing hallucinations. Retrieval-Augmented Generation (RAG) enhances LLM performance by integrating external knowledge bases, providing additional context which significantly improves accuracy and knowledge coverage. However, building these external knowledge bases often requires substantial resources and may involve sensitive information. In this paper, we propose an agent-based automated privacy attack called RAG-Thief, which can extract a scalable amount of private data from the private database used in RAG applications. We conduct a systematic study on the privacy risks associated with RAG applications, revealing that the vulnerability of LLMs makes the private knowledge bases suffer significant privacy risks. Unlike previous manual attacks which rely on traditional prompt injection techniques, RAG-Thief starts with an initial adversarial query and learns from model responses, progressively generating new queries to extract as many chunks from the knowledge base as possible. Experimental results show that our RAG-Thief can extract over 70% information from the private knowledge bases within customized RAG applications deployed on local machines and real-world platforms, including OpenAI's GPTs and ByteDance's Coze. Our findings highlight the privacy vulnerabilities in current RAG applications and underscore the pressing need for stronger safeguards.

Large language models (LLMs) deliver impressive results but face challenges from increasing model sizes and computational costs. Structured pruning reduces model size and speeds up inference but often causes uneven degradation across domains, leading to biased performance. To address this, we propose DRPruning, which incorporates distributionally robust optimization to restore balanced performance across domains, along with further improvements to enhance robustness. Experiments in monolingual and multilingual settings show that our method surpasses similarly sized models in pruning and continued pretraining over perplexity, downstream tasks, and instruction tuning. We further provide analysis demonstrating the robustness of our method towards various domains and distribution shifts. Furthermore, our method automatically determines optimal reference losses and data ratios, suggesting potential for broader applications. Our code is available at https://github.com/hexuandeng/DRPruning., Comment: Work in Progress

Multiple federated learning (FL) methods are proposed for traffic flow forecasting (TFF) to avoid heavy-transmission and privacy-leaking concerns resulting from the disclosure of raw data in centralized methods. However, these FL methods adopt offline learning which may yield subpar performance, when concept drift occurs, i.e., distributions of historical and future data vary. Online learning can detect concept drift during model training, thus more applicable to TFF. Nevertheless, the existing federated online learning method for TFF fails to efficiently solve the concept drift problem and causes tremendous computing and communication overhead. Therefore, we propose a novel method named Resource-Efficient Federated Online Learning (REFOL) for TFF, which guarantees prediction performance in a communication-lightweight and computation-efficient way. Specifically, we design a data-driven client participation mechanism to detect the occurrence of concept drift and determine clients' participation necessity. Subsequently, we propose an adaptive online optimization strategy, which guarantees prediction performance and meanwhile avoids meaningless model updates. Then, a graph convolution-based model aggregation mechanism is designed, aiming to assess participants' contribution based on spatial correlation without importing extra communication and computing consumption on clients. Finally, we conduct extensive experiments on real-world datasets to demonstrate the superiority of REFOL in terms of prediction improvement and resource economization.

Edge-AI, the convergence of edge computing and artificial intelligence (AI), has become a promising paradigm that enables the deployment of advanced AI models at the network edge, close to users. In Edge-AI, federated continual learning (FCL) has emerged as an imperative framework, which fuses knowledge from different clients while preserving data privacy and retaining knowledge from previous tasks as it learns new ones. By so doing, FCL aims to ensure stable and reliable performance of learning models in dynamic and distributed environments. In this survey, we thoroughly review the state-of-the-art research and present the first comprehensive survey of FCL for Edge-AI. We categorize FCL methods based on three task characteristics: federated class continual learning, federated domain continual learning, and federated task continual learning. For each category, an in-depth investigation and review of the representative methods are provided, covering background, challenges, problem formalisation, solutions, and limitations. Besides, existing real-world applications empowered by FCL are reviewed, indicating the current progress and potential of FCL in diverse application domains. Furthermore, we discuss and highlight several prospective research directions of FCL such as algorithm-hardware co-design for FCL and FCL with foundation models, which could provide insights into the future development and practical deployment of FCL in the era of Edge-AI.

The evolution of the lunar magnetic field can reveal the Moon's interior structure, thermal history, and surface environment. The mid-to-late stage evolution of the lunar magnetic field is poorly constrained, and thus the existence of a long-lived lunar dynamo remains controversial. The Chang'e-5 mission returned the heretofore youngest mare basalts from Oceanus Procellarum uniquely positioned at mid-latitude. We recovered weak paleointensities of 2-4 uT from the Chang'e-5 basalt clasts at 2 billion years ago, attestting to the longevity of a lunar dynamo until at least the Moon's midlife. This paleomagnetic result implies the existence of thermal convection in the lunar deep interior at the lunar mid-stage which may have supplied mantle heat flux for the young volcanism.

We propose Hymba, a family of small language models featuring a hybrid-head parallel architecture that integrates transformer attention mechanisms with state space models (SSMs) for enhanced efficiency. Attention heads provide high-resolution recall, while SSM heads enable efficient context summarization. Additionally, we introduce learnable meta tokens that are prepended to prompts, storing critical information and alleviating the "forced-to-attend" burden associated with attention mechanisms. This model is further optimized by incorporating cross-layer key-value (KV) sharing and partial sliding window attention, resulting in a compact cache size. During development, we conducted a controlled study comparing various architectures under identical settings and observed significant advantages of our proposed architecture. Notably, Hymba achieves state-of-the-art results for small LMs: Our Hymba-1.5B-Base model surpasses all sub-2B public models in performance and even outperforms Llama-3.2-3B with 1.32% higher average accuracy, an 11.67x cache size reduction, and 3.49x throughput., Comment: 20 pages, models are available on huggingface

The CMS and ATLAS Collaborations have recently reported their findings based on the comprehensive run 2 dataset, detailing their searches for a light Higgs boson with a mass of approximately 95 GeV. We investigate the excesses observed in the $\gamma\gamma$, $\tau{\bar\tau}$ and $b{\bar b}$ data at approximately 95 GeV in the $U(1)_X$ model. Additionally, it also mixes with the SM-like Higgs boson. Research indicates that, in this model, identifying this singlet Higgs state as the lightest Higgs boson holds tremendous potential for explaining the excess observed at approximately 95 GeV. In our calculations, we maintain the masses of the lightest and next-to-lightest Higgs bosons at approximately 95 GeV and 125 GeV, respectively. The study finds that the theoretical predictions for the signal strengths $\mu(h_{95})_{\gamma\gamma}$ $\mu(h_{95})_{\tau{\bar\tau}}$ and $\mu(h_{95})_{b{\bar b}}$ in the $U(1)_X$ model align well with the excesses observed by CMS., Comment: 23 pages, 5 figures

For any $a\in\mathbb{C}$, the zeros of $\zeta(s)-a$, denoted by $\rho_a=\beta_a+i\gamma_a$, are called $a$-points of the Riemann zeta function $\zeta(s)$. In this paper, we reformulate some basic results about the $a$-points of $\zeta(s)$ shown by Garunk\v{s}tis and Steuding. We then deduce an asymptotic of the sum \[S_T(a,\delta)=\sum_{\tau<\gamma_a\leqslant T}\zeta'(\rho_a+i\delta)X^{\rho_a},\quad T\to\infty,\] where $0\ne\delta=\frac{2\pi\alpha}{\log\frac{T}{2\pi X}}\ll 1$, and $X>0$ and $\tau\geqslant|\delta|+1$ are fixed. We also find the interesting varied behavior of $S_T(a,\delta)$ in different $X$ ranges, which is more complicated than those described before by Gonek and Pearce-Crump., Comment: 16 pages, 0 figures

Text-to-speech (TTS) models have been widely adopted to enhance automatic speech recognition (ASR) systems using text-only corpora, thereby reducing the cost of labeling real speech data. Existing research primarily utilizes additional text data and predefined speech styles supported by TTS models. In this paper, we propose Hard-Synth, a novel ASR data augmentation method that leverages large language models (LLMs) and advanced zero-shot TTS. Our approach employs LLMs to generate diverse in-domain text through rewriting, without relying on additional text data. Rather than using predefined speech styles, we introduce a hard prompt selection method with zero-shot TTS to clone speech styles that the ASR model finds challenging to recognize. Experiments demonstrate that Hard-Synth significantly enhances the Conformer model, achieving relative word error rate (WER) reductions of 6.5\%/4.4\% on LibriSpeech dev/test-other subsets. Additionally, we show that Hard-Synth is data-efficient and capable of reducing bias in ASR.

In this compelling study, we delve into the intricate relationship between the Equation of State (EoS) that encompasses a ``QGP-like'' phase transition and the directed flow ($v_1$) of protons and $\Lambda$ particles, specifically at beam energies under 10 GeV. Utilizing the sophisticated Multi-Phase Transport (AMPT-HC) model, our analysis not only resonates with the STAR experimental data but also reveals striking dips in the directed flow strength for both types of particles as the collision energy is ramped up. These pronounced features, which are more noticeable across various phase-transition regions within the EoS, offer invaluable insights into the presence and detailed nature of the QCD phase transition. Our findings underline the critical role of directed flow measurements in beam energy scans as effective instruments for probing and understanding the phase structure of nuclear matter in heavy-ion collisions, highlighting their potential to unlock new frontiers in high-energy nuclear physics., Comment: 5 pages, 3 figures, submitted

Recent literature reports a color deviation between observed Gaia color-magnitude diagrams (CMDs) and theoretical model isochrone predictions, particularly in the very low-mass regime. To assess its impact on cluster age determination via isochrone fitting, we quantified the color deviations for three benchmark clusters, Hyades, Pleiades, and Praesepe, both for the Gaia color (BP-RP) and (G-RP). In general, the (G-RP) color deviations are smaller than the (BP-RP) ones. Empirical color correction functions based on these benchmarks are derived for the currently available MIST and PARSEC 1.2S isochrone models. Applying the correction functions to 31 additional open clusters and 3 moving groups results in a significantly improved alignment between the isochrones and observed CMDs. With our empirical corrections, isochrones provide age estimates consistent with literature values obtained through the spectral Lithium Depletion Boundary method, validating the effectiveness of our approach. The corresponding metallicities with PARSEC 1.2S also show a good agreement with the spectroscopic results. The empirical color correction function we present in this work offers a tool for a consistent age determination within the full mass range of stellar clusters using the isochrone fitting method., Comment: 27 pages. Accepted for publication in ApJ

We study the performance guarantees of exploration-free greedy algorithms for the linear contextual bandit problem. We introduce a novel condition, named the \textit{Local Anti-Concentration} (LAC) condition, which enables a greedy bandit algorithm to achieve provable efficiency. We show that the LAC condition is satisfied by a broad class of distributions, including Gaussian, exponential, uniform, Cauchy, and Student's~$t$ distributions, along with other exponential family distributions and their truncated variants. This significantly expands the class of distributions under which greedy algorithms can perform efficiently. Under our proposed LAC condition, we prove that the cumulative expected regret of the greedy algorithm for the linear contextual bandit is bounded by $O(\operatorname{poly} \log T)$. Our results establish the widest range of distributions known to date that allow a sublinear regret bound for greedy algorithms, further achieving a sharp poly-logarithmic regret., Comment: NeurIPS2024

Interactive Medical Image Segmentation (IMIS) has long been constrained by the limited availability of large-scale, diverse, and densely annotated datasets, which hinders model generalization and consistent evaluation across different models. In this paper, we introduce the IMed-361M benchmark dataset, a significant advancement in general IMIS research. First, we collect and standardize over 6.4 million medical images and their corresponding ground truth masks from multiple data sources. Then, leveraging the strong object recognition capabilities of a vision foundational model, we automatically generated dense interactive masks for each image and ensured their quality through rigorous quality control and granularity management. Unlike previous datasets, which are limited by specific modalities or sparse annotations, IMed-361M spans 14 modalities and 204 segmentation targets, totaling 361 million masks-an average of 56 masks per image. Finally, we developed an IMIS baseline network on this dataset that supports high-quality mask generation through interactive inputs, including clicks, bounding boxes, text prompts, and their combinations. We evaluate its performance on medical image segmentation tasks from multiple perspectives, demonstrating superior accuracy and scalability compared to existing interactive segmentation models. To facilitate research on foundational models in medical computer vision, we release the IMed-361M and model at https://github.com/uni-medical/IMIS-Bench.

We use the Tannakian formalism to define the Emerton--Gee stack for general groups. For a flat algebraic group G over Z_p, we are able to prove the associated Emerton--Gee stack is a formal algebraic stack locally of finite presentation over Spf(Z_p). We also define a derived stack of Laurent F-crystals with G-structure on the absolute prismatic site, whose underlying classical stack is proved to be equivalent to the Emerton--Gee stack. In the case of connected reductive groups, we show that the derived stack of Laurent F-crystals with G-structure is classical in the sense that when restricted to truncated animated rings, it is the \'etale sheafification of the left Kan extension of the Emerton--Gee stack along the inclusion from classical commutative rings to animated rings. Moreover, when G is a generalised reductive group, the classicality result still holds for a modified version of the Emerton--Gee stack. In particular, this completes the picture that the derived stack of local Langlands parameters for the Langlands dual group of a reductive group is classical.

Misaligned circumbinary disks will produce dust traffic jams during alignment or anti-alignment to the binary orbital plane. We conduct a hydrodynamical simulation of an initially misaligned circumbinary disk undergoing polar alignment with multiple dust species. Due to differential precession between the gas and dust components, multiple dust traffic jams are produced within the disk during polar alignment. The radial locations of the dust traffic jams depend on the Stokes number of the grains, which depends on grain size. We compute the dust temperature structure using post-processing radiative transfer to produce continuum images at cm-wavelengths. Multiple distinct rings emerge in the continuum images, corresponding to the dust traffic jams. The angular resolution of upcoming observations from SKA and ngVLA will be sufficient to detect centimeter-sized grains in protoplanetary disks and resolve the widths of dust traffic jams. Therefore, dust traffic jams resulting from the differential precession of gas and dust in misaligned circumbinary disks will be a prime target for more extended wavelength observations., Comment: 12 pages, 5 figures, accepted for publication in ApJL

Combinatorial optimization (CO) is essential for improving efficiency and performance in engineering applications. As complexity increases with larger problem sizes and more intricate dependencies, identifying the optimal solution become challenging. When it comes to real-world engineering problems, algorithms based on pure mathematical reasoning are limited and incapable to capture the contextual nuances necessary for optimization. This study explores the potential of Large Language Models (LLMs) in solving engineering CO problems by leveraging their reasoning power and contextual knowledge. We propose a novel LLM-based framework that integrates network topology and domain knowledge to optimize the sequencing of Design Structure Matrix (DSM)-a common CO problem. Our experiments on various DSM cases demonstrate that the proposed method achieves faster convergence and higher solution quality than benchmark methods. Moreover, results show that incorporating contextual domain knowledge significantly improves performance despite the choice of LLMs. These findings highlight the potential of LLMs in tackling complex real-world CO problems by combining semantic and mathematical reasoning. This approach paves the way for a new paradigm in in real-world combinatorial optimization.

We calculate the leading-twist light-cone distribution amplitudes of the light $\Lambda$ baryon using lattice methods within the framework of large momentum effective theory. Our numerical computations are conducted employing $N_f=2+1$ stout smeared clover fermions and a Symanzik gauge action on a lattice with spacing $a=0.077\;\rm{fm}$, and a pion mass of 303 MeV. To approach the large momentum regime, we simulate the equal-time correlations with the hadron momentum $P^z = \{2.52, 3.02, 3.52\}$ GeV. By investigating the potential analytic characteristics of the baryon quasi-distribution amplitude in coordinate space, we validate these findings through our lattice calculations. After renormalization and extrapolation, we present results for the three-dimensional distribution of momentum fractions for the two light quarks. Based on these findings the paper briefly discusses the phenomenological impact on weak decays of $\Lambda_b$, and outlines potential systematic uncertainties that can be improved in the future. This work lays the theoretical foundation for accessing baryon LCDAs from lattice QCD.

Indoor SLAM often suffers from issues such as scene drifting, double walls, and blind spots, particularly in confined spaces with objects close to the sensors (e.g. LiDAR and cameras) in reconstruction tasks. Real-time visualization of point cloud registration during data collection may help mitigate these issues, but a significant limitation remains in the inability to in-depth compare the scanned data with actual physical environments. These challenges obstruct the quality of reconstruction products, frequently necessitating revisit and rescan efforts. For this regard, we developed the LiMRSF (LiDAR-MR-RGB Sensor Fusion) system, allowing users to perceive the in-situ point cloud registration by looking through a Mixed-Reality (MR) headset. This tailored framework visualizes point cloud meshes as holograms, seamlessly matching with the real-time scene on see-through glasses, and automatically highlights errors detected while they overlap. Such holographic elements are transmitted via a TCP server to an MR headset, where it is calibrated to align with the world coordinate, the physical location. This allows users to view the localized reconstruction product instantaneously, enabling them to quickly identify blind spots and errors, and take prompt action on-site. Our blind spot detector achieves an error detection precision with an F1 Score of 75.76% with acceptably high fidelity of monitoring through the LiMRSF system (highest SSIM of 0.5619, PSNR of 14.1004, and lowest MSE of 0.0389 in the five different sections of the simplified mesh model which users visualize through the LiMRSF device see-through glasses). This method ensures the creation of detailed, high-quality datasets for 3D models, with potential applications in Building Information Modeling (BIM) but not limited., Comment: 21 pages, 13 figures, 3 tables

It is widely acknowledged that the performance of Transformer models is exponentially related to their number of parameters and computational complexity. While approaches like Mixture of Experts (MoE) decouple parameter count from computational complexity, they still face challenges in inference due to high memory access costs. This work introduces UltraMem, incorporating large-scale, ultra-sparse memory layer to address these limitations. Our approach significantly reduces inference latency while maintaining model performance. We also investigate the scaling laws of this new architecture, demonstrating that it not only exhibits favorable scaling properties but outperforms traditional models. In our experiments, we train networks with up to 20 million memory slots. The results show that our method achieves state-of-the-art inference speed and model performance within a given computational budget., Comment: 10 pages, 6 figures

The opening or closing mechanism of a voltage-gated ion channel is triggered by the potential difference crossing the cell membrane in the nervous system. Based on this picture, we model the ion channel as a nanoscale two-terminal ionic tunneling junction. External time-varying voltage exerting on the two-terminal ionic tunneling junction mimics the stimulation of neurons from the outside. By deriving the quantum Langevin equation from quantum mechanics, the ion channel current is obtained by the quantum tunneling of ions controlled by the time-varying voltage. The time-varying voltage induces an effective magnetic flux which causes quantum coherence in ion tunnelings and leads to sideband effects in the ion channel current dynamics. The sideband effects in the ionic current dynamics manifest a multi-crossing hysteresis in the I-V curve, which is the memory dynamics responding to the variation of the external voltage. Such memory dynamics is defined as the active quantum memory with respect to the time-varying stimuli. We can quantitatively describe how active quantum memory is generated and changed. We find that the number of the non-zero cross points in the I-V curve hysteresis and the oscillation of the differential conductance are the characteristics for quantitatively describing the active quantum memory. We also explore the temperature dependence of the active quantum memory in such a system. The discovery of this active quantum memory characteristics provides a new understanding about the underlying mechanism of ion channel dynamics., Comment: 17 pages, 7 figures

Recently, test-time scaling has garnered significant attention from the research community, largely due to the substantial advancements of the o1 model released by OpenAI. By allocating more computational resources during the inference phase, large language models~(LLMs) can extensively explore the solution space by generating more thought tokens or diverse solutions, thereby producing more accurate responses. However, developing an o1-like reasoning approach is challenging, and researchers have been making various attempts to advance this open area of research. In this paper, we present a preliminary exploration into enhancing the reasoning abilities of LLMs through reward-guided tree search algorithms. This framework is implemented by integrating the policy model, reward model, and search algorithm. It is primarily constructed around a tree search algorithm, where the policy model navigates a dynamically expanding tree guided by a specially trained reward model. We thoroughly explore various design considerations necessary for implementing this framework and provide a detailed report of the technical aspects. To assess the effectiveness of our approach, we focus on mathematical reasoning tasks and conduct extensive evaluations on four challenging datasets, significantly enhancing the reasoning abilities of LLMs., Comment: LLM;Complex Reasoning;Math

Using $e^+e^-$ collision data corresponding to an integrated luminosity of 19 fb$^{-1}$ collected by the BESIII detector at center-of-mass energies ranging from 4.13 to 4.70 GeV, we report the first evidence for a new excited $\Omega^{-}$ hyperon, the $\Omega^*(2109)^{-}$, through the process $e^+ e^- \to \Omega^*(2109)^{-} \bar{\Omega}^{+} +c.c.$ with a significance of 3.7 $\sigma$. The mass and width of $\Omega^*(2109)^{-}$ are measured to be $2108.8 \pm 5.5_{\rm stat} \pm 1.5_{\rm syst} {\rm MeV}/c^{2}$ and $21.6 \pm 17.7_{\rm stat} \pm 9.4_{\rm syst} {\rm MeV}$, respectively. We also present evidence for production of the $\Omega^*(2012)^{-}$ in the process $e^+ e^- \to \Omega^*(2012)^{-} \bar{\Omega}^{+} +c.c.$ with a significance of 3.7 $\sigma$., Comment: 8 pages, 2 figures

Recent studies reveal that Large Language Models (LLMs) are susceptible to backdoor attacks, where adversaries embed hidden triggers that manipulate model responses. Existing backdoor defense methods are primarily designed for vision or classification tasks, and are thus ineffective for text generation tasks, leaving LLMs vulnerable. We introduce Internal Consistency Regularization (CROW), a novel defense using consistency regularization finetuning to address layer-wise inconsistencies caused by backdoor triggers. CROW leverages the intuition that clean models exhibit smooth, consistent transitions in hidden representations across layers, whereas backdoored models show noticeable fluctuation when triggered. By enforcing internal consistency through adversarial perturbations and regularization, CROW neutralizes backdoor effects without requiring clean reference models or prior trigger knowledge, relying only on a small set of clean data. This makes it practical for deployment across various LLM architectures. Experimental results demonstrate that CROW consistently achieves a significant reductions in attack success rates across diverse backdoor strategies and tasks, including negative sentiment, targeted refusal, and code injection, on models such as Llama-2 (7B, 13B), CodeLlama (7B, 13B) and Mistral-7B, while preserving the model's generative capabilities.

Nonlinear electrodynamics in materials manifests as an electronic response that depends on second- or higher-order powers of the applied electromagnetic field. This response is highly dependent on the underlying crystal symmetries in the material and is typically smaller than the linear responses. Nonlinear responses are therefore usually employed to expose the symmetry breaking, geometric properties of the electronic band structure in materials. Naturally, a material system with a strong nonlinear response is also the key component in nonlinear devices. Here we report the strong room-temperature second-harmonic transport response in a quantum magnet,TbMn6Sn6, which is governed by the quantum metric and can be tuned with applied magnetic fields and temperature. We show that around room temperature, which is close to the spontaneous spin-reorientation transition, the magnetic configurations, and therefore the related symmetry breaking phases, are easily controlled. Our results pave the way from quantum materials to high performance tuneable nonlinear device applications at room temperature., Comment: 12 pages, 3 figures

Recently, implicit neural representations (INRs) have attracted increasing attention for multi-dimensional data recovery. However, INRs simply map coordinates via a multi-layer perception (MLP) to corresponding values, ignoring the inherent semantic information of the data. To leverage semantic priors from the data, we propose a novel Superpixel-informed INR (S-INR). Specifically, we suggest utilizing generalized superpixel instead of pixel as an alternative basic unit of INR for multi-dimensional data (e.g., images and weather data). The coordinates of generalized superpixels are first fed into exclusive attention-based MLPs, and then the intermediate results interact with a shared dictionary matrix. The elaborately designed modules in S-INR allow us to ingenuously exploit the semantic information within and across generalized superpixels. Extensive experiments on various applications validate the effectiveness and efficacy of our S-INR compared to state-of-the-art INR methods., Comment: Accepted at ECCV 2024, 18 pages, 7 figures

Surgical instrument segmentation (SIS) is an essential task in computer-assisted surgeries, with deep learning-based research improving accuracy in complex environments. Recently, text-promptable segmentation methods have been introduced to generate masks based on text prompts describing target objects. However, these methods assume that the object described by a given text prompt exists in the scene. This results in mask generation whenever a related text prompt is provided, even if the object is absent from the image. Existing methods handle this by using prompts only for objects known to be present in the image, which introduces inaccessible information in a vision-based method setting and results in unfair comparisons. For fair comparison, we redefine existing text-promptable SIS settings to robust conditions, called Robust text-promptable SIS (R-SIS), designed to forward prompts of all classes and determine the existence of an object from a given text prompt for the fair comparison. Furthermore, we propose a novel framework, Robust Surgical Instrument Segmentation (RoSIS), which combines visual and language features for promptable segmentation in the R-SIS setting. RoSIS employs an encoder-decoder architecture with a Multi-Modal Fusion Block (MMFB) and a Selective Gate Block (SGB) to achieve balanced integration of vision and language features. Additionally, we introduce an iterative inference strategy that refines segmentation masks in two steps: an initial pass using name-based prompts, followed by a refinement step using location prompts. Experiments on various datasets and settings demonstrate that RoSIS outperforms existing vision-based and promptable methods under robust conditions., Comment: 10 pages, 6 figures, submitted to IEEE transactions on Medical Imaging

This paper presents the first analysis of the contact binary TYC 3801-1529-1. We observed four sets of multiple bands complete light curves and one set of radial velocity curve of the primary component. Based on a simultaneous investigation of our observed and TESS light curves and the radial velocity curve, we found that TYC 3801-1529-1 is an extremely low-mass-ratio, medium contact binary with $q=0.0356$, with the contribution of the third light at a level of about 10\%. Its mass ratio is lower than V1187 Her, making TYC 3801-1529-1 the lowest mass-ratio contact binary ever found in the universe. The light curves observed in 2022 are asymmetric, which is aptly explained by a hot spot on the primary component. A 16-year eclipse timings analysis indicates a secular increase orbital period with a rate of dp/dt$=7.96(\pm0.35)\times10^{-7}$ d yr$^{-1}$. We studied the stability of this target and identified that not only the value of $J_{spin}/J_{orb}$, but also the mass ratio surpass the unstable boundary. Hence, TYC 3801-1529-1 presents a challenge to theoretical research and ought to be considered a progenitor of a contact binary merger., Comment: 6 pages, 3 figures, and 1 table, accepted by A&A Letters, Data available via China-VO PaperData repository

Computing maximum/minimum distances between 3D meshes is crucial for various applications, i.e., robotics, CAD, VR/AR, etc. In this work, we introduce a highly parallel algorithm (gDist) optimized for Graphics Processing Units (GPUs), which is capable of computing the distance between two meshes with over 15 million triangles in less than 0.4 milliseconds (Fig. 1). By testing on benchmarks with varying characteristics, the algorithm achieves remarkable speedups over prior CPU-based and GPU-based algorithms on a commodity GPU (NVIDIA GeForce RTX 4090). Notably, the algorithm consistently maintains high-speed performance, even in challenging scenarios that pose difficulties for prior algorithms.

In this paper, we propose a new well-balanced fifth-order finite volume WENO method for solving one- and two-dimensional shallow water equations with bottom topography. The well-balanced property is crucial to the ability of a scheme to simulate perturbation waves over the ``lake-at-rest'' steady state such as waves on a lake or tsunami waves in the deep ocean. We adopt the constant subtraction technique such that both the flux gradient and source term in the new pre-balanced form vanish at the lake-at-rest steady state, while the well-balanced WENO method by Xing and Shu [Commun. Comput. Phys., 2006] uses high-order accurate numerical discretization of the source term and makes the exact balance between the source term and the flux gradient, to achieve the well-balanced property. The scaling positivity-preserving limiter is used for the water height near the dry areas. The fifth-order WENO-AO reconstruction is used to construct the solution since it has better resolution than the WENO-ZQ and WENO-MR reconstructions for the perturbation of steady state flows. Extensive one- and two-dimensional numerical examples are presented to demonstrate the well-balanced, fifth-order accuracy, non-oscillatory, and positivity-preserving properties of the proposed method., Comment: 38 pages, 17 figures

Artificial Intelligence (AI) has demonstrated significant capabilities in various fields, and in areas such as human-computer interaction (HCI), embodied intelligence, and the design and animation of virtual digital humans, both practitioners and users are increasingly concerned with AI's ability to understand and express emotion. Consequently, the question of whether AI can accurately interpret human emotions remains a critical challenge. To date, two primary classes of AI models have been involved in human emotion analysis: generative models and Multimodal Large Language Models (MLLMs). To assess the emotional capabilities of these two classes of models, this study introduces MEMO-Bench, a comprehensive benchmark consisting of 7,145 portraits, each depicting one of six different emotions, generated by 12 Text-to-Image (T2I) models. Unlike previous works, MEMO-Bench provides a framework for evaluating both T2I models and MLLMs in the context of sentiment analysis. Additionally, a progressive evaluation approach is employed, moving from coarse-grained to fine-grained metrics, to offer a more detailed and comprehensive assessment of the sentiment analysis capabilities of MLLMs. The experimental results demonstrate that existing T2I models are more effective at generating positive emotions than negative ones. Meanwhile, although MLLMs show a certain degree of effectiveness in distinguishing and recognizing human emotions, they fall short of human-level accuracy, particularly in fine-grained emotion analysis. The MEMO-Bench will be made publicly available to support further research in this area.

In this paper, we develop a simple, efficient, and fifth-order finite difference interpolation-based Hermite WENO (HWENO-I) scheme for one- and two-dimensional hyperbolic conservation laws. We directly interpolate the solution and first-order derivative values and evaluate the numerical fluxes based on these interpolated values. We do not need the split of the flux functions when reconstructing numerical fluxes and there is no need for any additional HWENO interpolation for the modified derivative. The HWENO interpolation only needs to be applied one time which utilizes the same candidate stencils, Hermite interpolation polynomials, and linear/nonlinear weights for the solution and first-order derivative at the cell interface, as well as the modified derivative at the cell center. The HWENO-I scheme inherits the advantages of the finite difference flux-reconstruction-based HWENO-R scheme [Fan et al., Comput. Methods Appl. Mech. Engrg., 2023], including fifth-order accuracy, compact stencils, arbitrary positive linear weights, and high resolution. The HWENO-I scheme is simpler and more efficient than the HWENO-R scheme and the previous finite difference interpolation-based HWENO scheme [Liu and Qiu, J. Sci. Comput., 2016] which needs the split of flux functions for the stability and upwind performance for the high-order derivative terms. Various benchmark numerical examples are presented to demonstrate the accuracy, efficiency, high resolution, and robustness of the proposed HWENO-I scheme., Comment: 36 pages, 15 figures

Based on 32162 molecular clouds from the Milky Way Imaging Scroll Painting project, we obtain new face-on molecular gas maps of the northern outer Galaxy. The total molecular gas surface density map reveals three segments of spirals, extending 16-43 kiloparsecs in length. The Perseus and Outer arms stand out prominently, appearing as quasi-continuous structures along most of their length. At the Galactic outskirts, about 1306 clouds connect the two segments of the new spiral arm discovered by Dame & Thaddeus (2011) in the first quadrant and Sun et al. (2015) in the second quadrant, possibly extending the arm into the outer third quadrant. Logarithmic spirals can be fitted to the CO arm segments with pitch angles ranging from 4 to 12 degree. These CO arms extend beyond previous CO studies and the optical radius, reaching a galactic radius of about 22 kiloparsecs, comparable to the HI radial range., Comment: 10 pages, 4 figures, 1 table, ApJL accepted

The accurate segmentation of retinal blood vessels plays a crucial role in the early diagnosis and treatment of various ophthalmic diseases. Designing a network model for this task requires meticulous tuning and extensive experimentation to handle the tiny and intertwined morphology of retinal blood vessels. To tackle this challenge, Neural Architecture Search (NAS) methods are developed to fully explore the space of potential network architectures and go after the most powerful one. Inspired by neuronal diversity which is the biological foundation of all kinds of intelligent behaviors in our brain, this paper introduces a novel and foundational approach to neural network design, termed ``neuron programming'', to automatically search neuronal types into a network to enhance a network's representation ability at the neuronal level, which is complementary to architecture-level enhancement done by NAS. Additionally, to mitigate the time and computational intensity of neuron programming, we develop a hypernetwork that leverages the search-derived architectural information to predict optimal neuronal configurations. Comprehensive experiments validate that neuron programming can achieve competitive performance in retinal blood segmentation, demonstrating the strong potential of neuronal diversity in medical image analysis.