Search

Since the peak of the pandemic in 2020, enrollment in virtual schools has steadily increased, with virtual schools now accounting for approximately 1.4% of the nation's public school students. While the effects of the pandemic on student achievement and mental health have been extensively studied, research has yet to thoroughly examine the impact of the pandemic on state policies and local organizational practices related to virtual schooling. Although some believed the onset of the pandemic would fundamentally reshape virtual schooling options in K-12 education, others were more skeptical, believing expanded virtual learning options were merely a temporary fix. This study examines whether the pandemic shaped virtual schooling at both the local and state levels, and if the changes made are likely to last and/or improve equity. The authors interviewed state and local education leaders and advocates, analyzed policy documents, websites, and media, observed legislative proceedings, and examined organizational practices in a sample of schools between 2019 and 2022 in Oregon. This state had an extensive history of virtual schooling before the pandemic, and enrollment continues to grow. The findings offer important lessons for policymakers and leaders to consider as the availability of virtual schools remains widespread across the country.

The physical origin of active galactic nucleus (AGN) variability remains unclear. Here we propose that the magnetic reconnection driven by the migration of satellite black holes (sBHs) in the AGN disc can be a new plausible mechanism for AGN short-term variability. During the sBH migration, the co-moving plasmas surrounding the sBH could influence the large-scale magnetic field of the AGN disk and trigger the magnetic reconnections to contribute to AGN UV/optical variability. Meanwhile, the plasma, which is accelerated by the magnetic reconnection, will successfully escape from the disc at Alfven velocity and cause a secondary magnetic reconnection in the corona. For a $\sim 10^{2}-10^{3}~{M_\mathrm{\odot}}$ sBH (including its binding gas) in the inner regions of the disc surrounding a supermassive black hole with $\sim 10^{7}~{M_\mathrm{\odot}}$, the reconnection process occurred in the space out of the disc can produce X-ray emission, which can last $\sim 10^4-10^6~\rm s$ with the luminosity $\sim 10^{39}- 10^{43}~\rm{erg ~s^{-1}}$., Comment: 7 pages, 2 figures, submitted

Transformers have emerged as the leading architecture in deep learning, proving to be versatile and highly effective across diverse domains beyond language and image processing. However, their impressive performance often incurs high computational costs due to their substantial model size. This paper focuses on compressing decoder-only transformer-based autoregressive models through structural weight pruning to improve the model efficiency while preserving performance for both language and image generation tasks. Specifically, we propose a training-free pruning method that calculates a numerical score with Newton's method for the Attention and MLP modules, respectively. Besides, we further propose another compensation algorithm to recover the pruned model for better performance. To verify the effectiveness of our method, we provide both theoretical support and extensive experiments. Our experiments show that our method achieves state-of-the-art performance with reduced memory usage and faster generation speeds on GPUs., Comment: Accepted by AAAI 2025

We introduce InternVL 2.5, an advanced multimodal large language model (MLLM) series that builds upon InternVL 2.0, maintaining its core model architecture while introducing significant enhancements in training and testing strategies as well as data quality. In this work, we delve into the relationship between model scaling and performance, systematically exploring the performance trends in vision encoders, language models, dataset sizes, and test-time configurations. Through extensive evaluations on a wide range of benchmarks, including multi-discipline reasoning, document understanding, multi-image / video understanding, real-world comprehension, multimodal hallucination detection, visual grounding, multilingual capabilities, and pure language processing, InternVL 2.5 exhibits competitive performance, rivaling leading commercial models such as GPT-4o and Claude-3.5-Sonnet. Notably, our model is the first open-source MLLMs to surpass 70% on the MMMU benchmark, achieving a 3.7-point improvement through Chain-of-Thought (CoT) reasoning and showcasing strong potential for test-time scaling. We hope this model contributes to the open-source community by setting new standards for developing and applying multimodal AI systems. HuggingFace demo see https://huggingface.co/spaces/OpenGVLab/InternVL, Comment: Technical Report

Large language models (LLMs) exhibit exceptional performance across various downstream tasks. However, they encounter limitations due to slow inference speeds stemming from their extensive parameters. The early exit (EE) is an approach that aims to accelerate auto-regressive decoding. EE generates outputs from intermediate layers instead of using the whole model, which offers a promising solution to this challenge. However, additional output layers and joint optimization used in conventional EE hinder the application of EE in LLMs. In this paper, we explore the possibility of LLMs EE without additional output layers and joint optimization. Our findings indicate that EE is a natural capability within transformer-based models. While joint optimization does not give model EE capability, it must be employed to address challenges by improving the accuracy of locating the optimal EE layer through gating functions. Additionally, our study reveals patterns in EE behavior from a sub-word perspective based on the LLaMA model and the potential possibility for EE based on sub-layers.

Residual networks, as discrete approximations of Ordinary Differential Equations (ODEs), have inspired significant advancements in neural network design, including multistep methods, high-order methods, and multi-particle dynamical systems. The precision of the solution to ODEs significantly affects parameter optimization, thereby impacting model performance. In this work, we present a series of advanced explorations of Transformer architecture design to minimize the error compared to the true ``solution.'' First, we introduce a predictor-corrector learning framework to minimize truncation errors, which consists of a high-order predictor and a multistep corrector. Second, we propose an exponential moving average-based coefficient learning method to strengthen our higher-order predictor. Extensive experiments on large-scale machine translation, abstractive summarization, language modeling, and natural language understanding benchmarks demonstrate the superiority of our approach. On the WMT'14 English-German and English-French tasks, our model achieved BLEU scores of 30.95 and 44.27, respectively. Furthermore, on the OPUS multilingual machine translation task, our model surpasses a robust 3.8B DeepNet by an average of 2.9 SacreBLEU, using only 1/3 parameters. Notably, it also beats LLama models by 5.7 accuracy points on the LM Harness Evaluation., Comment: Accepted by NeurIPS 2024

Large multimodal models (LMMs) have recently shown great progress in text-rich image understanding, yet they still struggle with complex, multi-page, visually-rich documents. Traditional methods using document parsers for retrieval-augmented generation suffer from performance and efficiency limitations, while directly presenting all pages to LMMs leads to inefficiencies, especially with lengthy documents. In this work, we present a novel framework named LoRA-Contextualizing Adaptation of Large multimodal models (LoCAL), which broadens the capabilities of any LMM to support long-document understanding. We demonstrate that LMMs can effectively serve as multimodal retrievers, fetching relevant pages to answer user questions based on these pages. LoCAL is implemented with two specific LMM adapters: one for evidence page retrieval and another for question answering. Empirical results show state-of-the-art performance on public benchmarks, demonstrating the effectiveness of LoCAL., Comment: Currently Under Review

Current popular Large Vision-Language Models (LVLMs) are suffering from Hallucinations on Object Attributes (HoOA), leading to incorrect determination of fine-grained attributes in the input images. Leveraging significant advancements in 3D generation from a single image, this paper proposes a novel method to mitigate HoOA in LVLMs. This method utilizes multiview images sampled from generated 3D representations as visual prompts for LVLMs, thereby providing more visual information from other viewpoints. Furthermore, we observe the input order of multiple multiview images significantly affects the performance of LVLMs. Consequently, we have devised Multiview Image Augmented VLM (MIAVLM), incorporating a Multiview Attributes Perceiver (MAP) submodule capable of simultaneously eliminating the influence of input image order and aligning visual information from multiview images with Large Language Models (LLMs). Besides, we designed and employed negative instructions to mitigate LVLMs' bias towards ``Yes" responses. Comprehensive experiments demonstrate the effectiveness of our method., Comment: 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2025)

Dynamic graph augmentation is used to improve the performance of dynamic GNNs. Most methods assume temporal locality, meaning that recent edges are more influential than earlier edges. However, for temporal changes in edges caused by random noise, overemphasizing recent edges while neglecting earlier ones may lead to the model capturing noise. To address this issue, we propose STAA (SpatioTemporal Activity-Aware Random Walk Diffusion). STAA identifies nodes likely to have noisy edges in spatiotemporal dimensions. Spatially, it analyzes critical topological positions through graph wavelet coefficients. Temporally, it analyzes edge evolution through graph wavelet coefficient change rates. Then, random walks are used to reduce the weights of noisy edges, deriving a diffusion matrix containing spatiotemporal information as an augmented adjacency matrix for dynamic GNN learning. Experiments on multiple datasets show that STAA outperforms other dynamic graph augmentation methods in node classification and link prediction tasks., Comment: 2025 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2025)

Estimating the shortest travel time and providing route recommendation between different locations in a city or region can quantitatively measure the conditions of the transportation network during or after extreme events. One common approach is to use Dijkstra's Algorithm, which produces the shortest path as well as the shortest distance. However, this option is computationally expensive when applied to large-scale networks. This paper proposes a novel fast framework based on graph neural networks (GNNs) which approximate the single-source shortest distance between pairs of locations, and predict the single-source shortest path subsequently. We conduct multiple experiments on synthetic graphs of different size to demonstrate the feasibility and computational efficiency of the proposed model. In real-world case studies, we also applied the proposed method of flood risk analysis of coastal urban areas to calculate delays in evacuation to public shelters during hurricanes. The results indicate the accuracy and computational efficiency of the GNN model, and its potential for effective implementation in emergency planning and management., Comment: 17 pages, 11 figures

Generative models have made significant impacts across various domains, largely due to their ability to scale during training by increasing data, computational resources, and model size, a phenomenon characterized by the scaling laws. Recent research has begun to explore inference-time scaling behavior in Large Language Models (LLMs), revealing how performance can further improve with additional computation during inference. Unlike LLMs, diffusion models inherently possess the flexibility to adjust inference-time computation via the number of denoising steps, although the performance gains typically flatten after a few dozen. In this work, we explore the inference-time scaling behavior of diffusion models beyond increasing denoising steps and investigate how the generation performance can further improve with increased computation. Specifically, we consider a search problem aimed at identifying better noises for the diffusion sampling process. We structure the design space along two axes: the verifiers used to provide feedback, and the algorithms used to find better noise candidates. Through extensive experiments on class-conditioned and text-conditioned image generation benchmarks, our findings reveal that increasing inference-time compute leads to substantial improvements in the quality of samples generated by diffusion models, and with the complicated nature of images, combinations of the components in the framework can be specifically chosen to conform with different application scenario.

We employ the SU(n)_k Wess-Zumino-Witten (WZW) model in conformal field theory to construct lattice wave functions in both one and two dimensions. It is unveiled that these wave functions can be reinterpreted as parton states, which enables efficient conversion to matrix product states such that many physical properties can be evaluated directly. In one dimension, these wave functions describe critical spin chains whose universality classes are in one-to-one correspondence with the WZW models used in the construction. In two dimensions, our constructions yield model wave functions for chiral spin liquids, and we show how to find all topological sectors of them in a systematic way. Using the null vectors of Kac-Moody algebras, parent Hamiltonians of the SU(3)_k series are derived. The SU(3)_k chiral spin liquids are lattice analogs of non-Abelian spin-singlet fractional quantum Hall states, and the k = 2 member hosts Fibonacci anyons., Comment: 14 pages, 6 figures

The origin of the correlation between the effective spins ($\chi_{\rm eff}$) and mass ratios ($q$) of LIGO-Virgo-KAGRA's binary black holes (BBHs) is still an open question. Motivated by recent identification of two subpopulations of the BBHs, in this work we investigate the potential $\chi_{\rm eff}-q$ correlation for each subpopulation. Surprisingly, the $\chi_{\rm eff}$-$q$ correlation vanishes for the low-mass subpopulation if we introduce a second $\chi_{\rm eff}$ distribution for the high-mass subpopulation likely originating from hierarchical mergers. The first subpopulation has a narrow $\chi_{\rm eff}$ distribution peaking at $\sim0.05$, whose primary-mass function cuts off at $\sim 45M_{\odot}$, in agreement with first-generation BBHs. The second $\chi_{\rm eff}$ distribution is broad and peaks at $\mu_{\chi,2}=0.35^{+0.18}_{-0.22}$, consistent with the expectation of hierarchical mergers formed in the disks of active galactic nucleus (AGNs). We infer $\mu_{\chi,2}>0$ at 98.7\% credible level, and a symmetric $\chi_{\rm eff}$ distribution for the second subpopulation is disfavored by $\mathcal{B}\sim5$. However, negative values of $\chi_{\rm eff}$ are also measured, indicating that the hierarchical mergers may take place within both star clusters and AGN disks. We find a Bayes factor of $\ln\mathcal{B}=5.2$ for two distinct $\chi_{\rm eff}$ distributions relative to single $\chi_{\rm eff}$ distribution that conditioned on mass ratios. Therefore we conclude that the $\chi_{\rm eff}$-$q$ correlation in the entire population can be explained as the superposition of two subpopulations. Additionally, we suggest to use a flexible mass function to reduce the bias in $\chi_{\rm eff}$-$q$ correlation that may be introduced by model mis-specification., Comment: 13 pages, 7 figures, comments welcome!

Despite significant advancements in general-purpose AI agents, several challenges still hinder their practical application in real-world scenarios. First, the limited planning capabilities of Large Language Models (LLM) restrict AI agents from effectively solving complex tasks that require long-horizon planning. Second, general-purpose AI agents struggle to efficiently utilize domain-specific knowledge and human expertise. In this paper, we introduce the Standard Operational Procedure-guided Agent (SOP-agent), a novel framework for constructing domain-specific agents through pseudocode-style Standard Operational Procedures (SOPs) written in natural language. Formally, we represent a SOP as a decision graph, which is traversed to guide the agent in completing tasks specified by the SOP. We conduct extensive experiments across tasks in multiple domains, including decision-making, search and reasoning, code generation, data cleaning, and grounded customer service. The SOP-agent demonstrates excellent versatility, achieving performance superior to general-purpose agent frameworks and comparable to domain-specific agent systems. Additionally, we introduce the Grounded Customer Service Benchmark, the first benchmark designed to evaluate the grounded decision-making capabilities of AI agents in customer service scenarios based on SOPs., Comment: 35 pages, 5 figures

Solving traffic assignment problem for large networks is computationally challenging when conventional optimization-based methods are used. In our research, we develop an innovative surrogate model for a traffic assignment when multi-class vehicles are involved. We do so by employing heterogeneous graph neural networks which use a multiple-view graph attention mechanism tailored to different vehicle classes, along with additional links connecting origin-destination pairs. We also integrate the node-based flow conservation law into the loss function. As a result, our model adheres to flow conservation while delivering highly accurate predictions for link flows and utilization ratios. Through numerical experiments conducted on urban transportation networks, we demonstrate that our model surpasses traditional neural network approaches in convergence speed and predictive accuracy in both user equilibrium and system optimal versions of traffic assignment., Comment: 16 pages, 5 figures

Chiral perovskites are emerging semiconducting materials with broken symmetry that can selectively absorb and emit circularly polarized light. However, most of the chiral perovskites are typically low-dimensional structures with limited electrical conductivity and their light absorption occurs in the UV region. In this work, we find doping 2,3,5,6-Tetrafluoro-7,7,8,8-tetracyanoquinodimethane (F4TCNQ) in the chiral perovskite matrix can improve the electrical conductivity with an addition of visible light absorption through the emerging charge-transfer electronic states. The new absorption feature exhibits strong circular dichroism adapted from the chiral matrix, which is indicative of a chirality transfer from the host to the guest via an electronic coupling. The charge transfer state is validated by transient absorption spectroscopy and theory modeling. Quantum-chemical modeling identifies a strong wave function overlap between an electron and a hole of the guest-host in a closely packed crystal configuration forming the charge-transfer absorption state. We then integrate the doped chiral perovskite film in photodetectors and demonstrate a selective detection of circularly polarized light both in the UV and visible range. Our results suggest a universal approach of introducing visible photo absorption states to the chiral matrix to broaden the optical active range and enhance the conductivity.

Hot carriers, inheriting excess kinetic energy from high-energy photons, underpin numerous optoelectronic applications involving non-equilibrium transport processes. Current research on hot carriers has predominantly focused on inorganic materials, with little attention paid to organic-based systems due to their ultrafast energy relaxation and inefficient charge transport. Here, we overturn this paradigm by demonstrating highly mobile hot carriers in solution-processable, highly crystalline two-dimensional conjugated coordination polymer (2D c-CP) Cu3BHT (BHT = benzenehexathiol) films. Leveraging a suite of ultrafast spectroscopic and imaging techniques, we unravel the microscopic charge transport landscape in Cu3BHT films following non-equilibrium photoexcitation across temporal, spatial, and frequency domains, revealing two distinct high-mobility transport regimes. In the non-equilibrium transport regime, hot carriers achieve ultrahigh mobility of ~2,000 cm2 V-1 s-1, traversing grain boundaries up to 300 nm within a picosecond. In the quasi-equilibrium transport regime, free carriers exhibit Drude-type band-like transport with a remarkable mobility of ~400 cm2 V-1 s-1 and an intrinsic diffusion length exceeding 1 micrometer. These findings establish 2D c-CPs as versatile platforms for exploring high-mobility non-equilibrium transport, unlocking new opportunities for organic-based hot carrier applications., Comment: 27 pages, 4 figures

Mental health is a critical global public health issue, and psychological support hotlines play a pivotal role in providing mental health assistance and identifying suicide risks at an early stage. However, the emotional expressions conveyed during these calls remain underexplored in current research. This study introduces a method that combines pitch acoustic features with deep learning-based features to analyze and understand emotions expressed during hotline interactions. Using data from China's largest psychological support hotline, our method achieved an F1-score of 79.13% for negative binary emotion classification.Additionally, the proposed approach was validated on an open dataset for multi-class emotion classification,where it demonstrated better performance compared to the state-of-the-art methods. To explore its clinical relevance, we applied the model to analysis the frequency of negative emotions and the rate of emotional change in the conversation, comparing 46 subjects with suicidal behavior to those without. While the suicidal group exhibited more frequent emotional changes than the non-suicidal group, the difference was not statistically significant.Importantly, our findings suggest that emotional fluctuation intensity and frequency could serve as novel features for psychological assessment scales and suicide risk prediction.The proposed method provides valuable insights into emotional dynamics and has the potential to advance early intervention and improve suicide prevention strategies through integration with clinical tools and assessments The source code is publicly available at https://github.com/Sco-field/Speechemotionrecognition/tree/main.

Different from traditional offline channel modeling, digital twin online channel modeling can sense and accurately characterize dynamic wireless channels in real time, and can therefore greatly assist 6G network optimization. This article proposes a novel promising framework and a step-by-step design procedure of digital twin online channel models (DTOCM). By enabling continuous visualization and accurate prediction of dynamic channel variations, DTOCM can synchronize the performance between simulated and real networks. We first explore the evolution and conceptual advancements of DTOCM, highlighting its visions and associated challenges. Then, we explain its operational principles, construction mechanisms, and applications to typical 6G scenarios. Subsequently, the real-time channel information provisioning and visualization capabilities of DTOCM are illustrated through our DTOCM platform based on practical scenarios. Finally, future research directions and open issues are discussed.

In recent years, 3D Gaussian splatting (3D-GS) has emerged as a novel scene representation approach. However, existing vision-only 3D-GS methods often rely on hand-crafted heuristics for point-cloud densification and face challenges in handling occlusions and high GPU memory and computation consumption. LiDAR-Inertial-Visual (LIV) sensor configuration has demonstrated superior performance in localization and dense mapping by leveraging complementary sensing characteristics: rich texture information from cameras, precise geometric measurements from LiDAR, and high-frequency motion data from IMU. Inspired by this, we propose a novel real-time Gaussian-based simultaneous localization and mapping (SLAM) system. Our map system comprises a global Gaussian map and a sliding window of Gaussians, along with an IESKF-based odometry. The global Gaussian map consists of hash-indexed voxels organized in a recursive octree, effectively covering sparse spatial volumes while adapting to different levels of detail and scales. The Gaussian map is initialized through multi-sensor fusion and optimized with photometric gradients. Our system incrementally maintains a sliding window of Gaussians, significantly reducing GPU computation and memory consumption by only optimizing the map within the sliding window. Moreover, we implement a tightly coupled multi-sensor fusion odometry with an iterative error state Kalman filter (IESKF), leveraging real-time updating and rendering of the Gaussian map. Our system represents the first real-time Gaussian-based SLAM framework deployable on resource-constrained embedded systems, demonstrated on the NVIDIA Jetson Orin NX platform. The framework achieves real-time performance while maintaining robust multi-sensor fusion capabilities. All implementation algorithms, hardware designs, and CAD models will be publicly available.

This is a book about large language models. As indicated by the title, it primarily focuses on foundational concepts rather than comprehensive coverage of all cutting-edge technologies. The book is structured into four main chapters, each exploring a key area: pre-training, generative models, prompting techniques, and alignment methods. It is intended for college students, professionals, and practitioners in natural language processing and related fields, and can serve as a reference for anyone interested in large language models.

Variational mode decomposition (VMD) and its extensions like Multivariate VMD (MVMD) decompose signals into ensembles of band-limited modes with narrow central frequencies. These methods utilize Fourier transformations to shift signals between time and frequency domains. However, since Fourier transformations span the entire time-domain signal, they are suboptimal for non-stationary time series. We introduce Short-Time Variational Mode Decomposition (STVMD), an innovative extension of the VMD algorithm that incorporates the Short-Time Fourier transform (STFT) to minimize the impact of local disturbances. STVMD segments signals into short time windows, converting these segments into the frequency domain. It then formulates a variational optimization problem to extract band-limited modes representing the windowed data. The optimization aims to minimize the sum of the bandwidths of these modes across the windowed data, extending the cost functions used in VMD and MVMD. Solutions are derived using the alternating direction method of multipliers, ensuring the extraction of modes with narrow bandwidths. STVMD is divided into dynamic and non-dynamic types, depending on whether the central frequencies vary with time. Our experiments show that non-dynamic STVMD is comparable to VMD with properly sized time windows, while dynamic STVMD better accommodates non-stationary signals, evidenced by reduced mode function errors and tracking of dynamic central frequencies. This effectiveness is validated by steady-state visual-evoked potentials in electroencephalogram signals., Comment: 13 pages, 11 figures

This article extends the Cox-Reid local parameter orthogonality to a multivariate setting and shows that the extension can lead to the Whittle algorithm for multivariate autoregressive models.

Recent advancements have highlighted the efficacy of self-supervised learning (SSL) features in various speech-related tasks, providing lightweight and versatile multi-view speech representations. However, our study reveals that while SSL features expedite model convergence, they conflict with traditional spectral features like FBanks in terms of update directions. In response, we propose a novel generalized feature fusion framework grounded in conditional computation, featuring a gradient-sensitive gating network and a multi-stage dropout strategy. This framework mitigates feature conflicts and bolsters model robustness to multi-view input features. By integrating SSL and spectral features, our approach accelerates convergence and maintains performance on par with spectral models across multiple speech translation tasks on the MUSTC dataset., Comment: ICASSP 2025

This article investigates the adaptive resource allocation scheme for digital twin (DT) synchronization optimization over dynamic wireless networks. In our considered model, a base station (BS) continuously collects factory physical object state data from wireless devices to build a real-time virtual DT system for factory event analysis. Due to continuous data transmission, maintaining DT synchronization must use extensive wireless resources. To address this issue, a subset of devices is selected to transmit their sensing data, and resource block (RB) allocation is optimized. This problem is formulated as a constrained Markov process (CMDP) problem that minimizes the long-term mismatch between the physical and virtual systems. To solve this CMDP, we first transform the problem into a dual problem that refines RB constraint impacts on device scheduling strategies. We then propose a continual reinforcement learning (CRL) algorithm to solve the dual problem. The CRL algorithm learns a stable policy across historical experiences for quick adaptation to dynamics in physical states and network capacity. Simulation results show that the CRL can adapt quickly to network capacity changes and reduce normalized root mean square error (NRMSE) between physical and virtual states by up to 55.2%, using the same RB number as traditional methods., Comment: 14 pages, 11 figures

The James Webb Space Telescope (JWST) has recently conducted observations of massive galaxies at high redshifts, revealing a notable anomaly in their star formation efficiency (SFE). Motivated by the recent identification of three $\sim 10^{6}M_\odot$ dark star candidates, we investigate whether dark stars can be the origin of the SFE excess. It turns out that the excess can be reproduced by a group of dark stars with $M \gtrsim 10^{3}\, \rm M_{\odot}$, because of their domination in generating primary UV radiation in high-redshift galaxies. The genesis of these dark stars is attributed to the capture of Weakly Interacting Massive Particles (WIMPs) within a mass range of tens of GeV to a few TeV. However, if the top-heavy initial mass function of dark stars holds up to $\sim 10^{5}M_\odot$, the relic black holes stemming from their collapse would be too abundant to be consistent with the current observations of Massive Compact Halo Objects (MACHOs). We thus suggest that just a small fraction of SFE excess may be contributed by the very massive dark stars and the majority likely originated from other reasons such as the Population III stars in view of their rather similar UV radiation efficiencies., Comment: Accepted for publication in ApJ

Previous work on augmenting large multimodal models (LMMs) for text-to-image (T2I) generation has focused on enriching the input space of in-context learning (ICL). This includes providing a few demonstrations and optimizing image descriptions to be more detailed and logical. However, as demand for more complex and flexible image descriptions grows, enhancing comprehension of input text within the ICL paradigm remains a critical yet underexplored area. In this work, we extend this line of research by constructing parallel multilingual prompts aimed at harnessing the multilingual capabilities of LMMs. More specifically, we translate the input text into several languages and provide the models with both the original text and the translations. Experiments on two LMMs across 3 benchmarks show that our method, PMT2I, achieves superior performance in general, compositional, and fine-grained assessments, especially in human preference alignment. Additionally, with its advantage of generating more diverse images, PMT2I significantly outperforms baseline prompts when incorporated with reranking methods. Our code and parallel multilingual data can be found at https://github.com/takagi97/PMT2I., Comment: Accepted to ICASSP 2025

Antiferromagnetism, characterized by magnetic moments aligned in alternating directions with a vanished ensemble average, has garnered renewed interest for its potential applications in spintronics and axion dynamics. The synergy between antiferromagnetism and topology can lead to the emergence of an exotic topological phase unique to certain magnetic order, termed antiferromagnetic topological insulators (AF TIs). A hallmark signature of AF TIs is the presence of a single surface Dirac cone--a feature typically associated with strong three-dimensional (3D) topological insulators--only on certain symmetry-preserving crystal terminations. However, the direct observation of this phenomenon poses a significant challenge. Here, we have theoretically and experimentally discovered a 3D photonic AF TI hosting a single surface Dirac cone protected by the combined symmetry of time reversal and half-lattice translation. Conceptually, our setup can be viewed as a z-directional stack of two-dimensional Chern insulators, with adjacent layers oppositely magnetized to form a 3D type-A AF configuration. By measuring both bulk and surface states, we have directly observed the symmetry-protected gapless single-Dirac-cone surface state, which shows remarkable robustness against random magnetic disorders. Our work constitutes the first realization of photonic AF TIs and photonic analogs of strong topological insulators, opening a new chapter for exploring novel topological photonic devices and phenomena that incorporate additional magnetic degrees of freedom., Comment: 13 pages, 4 figures

Cloud-based services are making the outsourcing of sensitive client data increasingly common. Although homomorphic encryption (HE) offers strong privacy guarantee, it requires substantially more resources than computing on plaintext, often leading to unacceptably large latencies in getting the results. HE accelerators have emerged to mitigate this latency issue, but with the high cost of ASICs. In this paper we show that HE primitives can be converted to AI operators and accelerated on existing ASIC AI accelerators, like TPUs, which are already widely deployed in the cloud. Adapting such accelerators for HE requires (1) supporting modular multiplication, (2) high-precision arithmetic in software, and (3) efficient mapping on matrix engines. We introduce the CROSS compiler (1) to adopt Barrett reduction to provide modular reduction support using multiplier and adder, (2) Basis Aligned Transformation (BAT) to convert high-precision multiplication as low-precision matrix-vector multiplication, (3) Matrix Aligned Transformation (MAT) to covert vectorized modular operation with reduction into matrix multiplication that can be efficiently processed on 2D spatial matrix engine. Our evaluation of CROSS on a Google TPUv4 demonstrates significant performance improvements, with up to 161x and 5x speedup compared to the previous work on many-core CPUs and V100. The kernel-level codes are open-sourced at https://github.com/google/jaxite.git., Comment: 16 pages, 10 figures, 4 algorithms, 7 tables. Enabling Google TPUv4 for privacy-preserving AI inference

The stochastic gravitational wave background (SGWB) can be observed in the nanohertz band using a pulsar timing array (PTA). Here a computationally efficient state-space framework is developed for analysing SGWB data, in which the stochastic gravitational wave strain at Earth is tracked with a non-linear Kalman filter and separated simultaneously from intrinsic, achromatic pulsar spin wandering. The filter is combined with a nested sampler to estimate the parameters of the model, and to calculate a Bayes factor for selecting between models with and without a SGWB. The procedure extends previous state-space formulations of PTA data analysis applied to individually resolvable binary black hole sources. The performance of the new algorithm is tested on synthetic data from the first International PTA Mock Data Challenge. It is shown that the algorithm distinguishes a SGWB from pure noise for $A_{\rm gw} \geq 3 \times 10^{-14}$, where $A_{\rm gw}$ denotes the standard normalization factor for a power spectral density with power-law exponent $-13/3$. Additional, systematic validation tests are also performed with synthetic data generated independently by adjusting the injected parameters to cover astrophysically plausible ranges. Full posterior distributions are recovered and tested for accuracy. The state-space procedure is memory-light and evaluates the likelihood for a standard-sized PTA dataset in $\lesssim 10^{-1}$ s without optimization on a standard central processing unit., Comment: 10 pages, 4 figures + appendices. Accepted for publication in MNRAS

As a fundamental QED process, linear Breit-Wheeler (LBW) pair production predicted 90 years ago has not yet been demonstrated in experiments with real photons. Here, we propose an experimentally advantageous scheme to detect the LBW signal by irradiating a foil target with a single 10 PW-level laser. Our integrated QED particle-in-cell simulations demonstrate that the LBW signal can be explicitly distinguished from the Bethe-Heitler (BH) signal by comparing positron energy spectra behind the target at varying target thicknesses. The LBW positrons are created at the front of the target and subsequently experience both laser vacuum acceleration and sheath field acceleration to gain high energies, while BH positrons, originating within the target bulk, are only subjected to sheath field acceleration. As a result, the invariance of the high-energy tail of positron spectra with respect to the target thickness serves as a distinct signature of the LBW process. Notably, this scheme remains viable even when the BH yield dominates over the LBW yield., Comment: 6 pages, 4 figures

This paper explores the synergy between human cognition and Large Language Models (LLMs), highlighting how generative AI can drive personalized learning at scale. We discuss parallels between LLMs and human cognition, emphasizing both the promise and new perspectives on integrating AI systems into education. After examining challenges in aligning technology with pedagogy, we review AutoTutor-one of the earliest Intelligent Tutoring Systems (ITS)-and detail its successes, limitations, and unfulfilled aspirations. We then introduce the Socratic Playground, a next-generation ITS that uses advanced transformer-based models to overcome AutoTutor's constraints and provide personalized, adaptive tutoring. To illustrate its evolving capabilities, we present a JSON-based tutoring prompt that systematically guides learner reflection while tracking misconceptions. Throughout, we underscore the importance of placing pedagogy at the forefront, ensuring that technology's power is harnessed to enhance teaching and learning rather than overshadow it.

Tucker decomposition has been widely used in a variety of applications to obtain latent factors of tensor data. In these applications, a common need is to compute Tucker decomposition for a given time range. Furthermore, real-world tensor time series are typically evolving in the time dimension. Such needs call for a data structure that can efficiently and accurately support range queries of Tucker decomposition and stream updates. Unfortunately, existing methods do not support either range queries or stream updates. This challenging problem has remained open for years prior to our work. To solve this challenging problem, we propose TUCKET, a data structure that can efficiently and accurately handle both range queries and stream updates. Our key idea is to design a new data structure that we call a stream segment tree by generalizing the segment tree, a data structure that was originally invented for computational geometry. For a range query of length $L$, our TUCKET can find $O(\log L)$ nodes (called the hit set) from the tree and efficiently stitch their preprocessed decompositions to answer the range query. We also propose an algorithm to optimally prune the hit set via an approximation of subtensor decomposition. For the $T$-th stream update, our TUCKET modifies only amortized $O(1)$ nodes and only $O(\log T)$ nodes in the worst case. Extensive evaluation demonstrates that our TUCKET consistently achieves the highest efficiency and accuracy across four large-scale datasets. Our TUCKET achieves at least 3 times lower latency and at least 1.4 times smaller reconstruction error than Zoom-Tucker on all datasets., Comment: Accepted at VLDB 2025

Efficient preference optimization algorithms such as Direct Preference Optimization (DPO) have become a popular approach in aligning large language models (LLMs) with human preferences. These algorithms implicitly treat the LLM as a reward model, and focus on training it to correct misranked preference pairs. However, recent work~\citep{chen2024preference} empirically finds that DPO training \textit{rarely improves these misranked preference pairs}, despite its gradient emphasizing on these cases. We introduce FocalPO, a DPO variant that instead \textit{down-weighs} misranked preference pairs and prioritizes enhancing the model's understanding of pairs that it can already rank correctly. Inspired by Focal Loss used in vision tasks, FocalPO achieves this by adding a modulating factor to dynamically scale DPO loss. Our experiment demonstrates that FocalPO surpasses DPO and its variants on popular benchmarks like Alpaca Eval 2.0 using Mistral-Base-7B and Llama-3-Instruct-8B. Additionally, we empirically reveals how FocalPO affects training on correct and incorrect sample groups, further underscoring its effectiveness.

A study is presented of $B^+ \to K^0_{\rm S} K^- \pi^+ K^-$ and $B^+ \to K^0_{\rm S} K^+ \pi^- K^+$ decays based on the analysis of proton-proton collision data collected with the LHCb detector at centre-of-mass energies of 7, 8 and 13 TeV, corresponding to an integrated luminosity of $9 fb^{-1}$. The $K^0_{\rm S} K \pi$ invariant-mass distributions of both $B^+$ decay modes show, in the $m(K^0_{\rm S} K \pi)<1.85$ GeV mass region, a rich spectrum of light-meson resonances, resolved using an amplitude analysis. A complex mixture of $J^{PC}=0^{-+}, 1^{++}$ and $1^{+-}$ resonances is observed, dominated by $\eta(1405)$, $\eta(1470)$, $\eta(1760)$, $f_1(1285)$, $f_1(1420)$ and $h_1(1405)$ resonances. The $K^0_{\rm S} K \pi$ Dalitz plots are dominated by asymmetric crossing $K^* \bar K$ bands which are different for the two $B^+$ decay modes. This is due to a different interference pattern between the $1^{++}$ and $1^{+-}$ amplitudes in the two channels. Branching fractions are measured for each resonant contribution., Comment: All figures and tables, along with any supplementary material and additional information, are available at https://cern.ch/lhcbproject/Publications/p/LHCb-PAPER-2024-045.html (LHCb public pages)

Figures and their captions play a key role in scientific publications. However, despite their importance, many captions in published papers are poorly crafted, largely due to a lack of attention by paper authors. While prior AI research has explored caption generation, it has mainly focused on reader-centered use cases, where users evaluate generated captions rather than actively integrating them into their writing. This paper addresses this gap by investigating how paper authors incorporate AI-generated captions into their writing process through a user study involving 18 participants. Each participant rewrote captions for two figures from their own recently published work, using captions generated by state-of-the-art AI models as a resource. By analyzing video recordings of the writing process through interaction analysis, we observed that participants often began by copying and refining AI-generated captions. Paper writers favored longer, detail-rich captions that integrated textual and visual elements but found current AI models less effective for complex figures. These findings highlight the nuanced and diverse nature of figure caption composition, revealing design opportunities for AI systems to better support the challenges of academic writing., Comment: This paper will appear at AAAI 2025 Workshop (2nd AI4Research Workshop: Towards a Knowledge-grounded Scientific Research Lifecycle)

Long-range magnetic ordering of the orbital motion of oxygen atoms within NO$_3$$^-$ cations is identified from experimental measurements of the magnetic susceptibility $\chi$($T$) in LiNO$_3$, Ca(NO$_3$)$_2$, K$_{0.5}$Rb$_{0.5}$NO$_3$ and C(NH$_2$)$_3$NO$_3$ at their respective order-disorder, solid-solid phase transitions $T$$_N$. The observed sharp changes in $\chi$($T$) and accompanying hysteretic behavior indicate the phase transitions to be first order. A model employing the law of conservation of angular momentum is used to explain why the librations between neighboring NO$_3$$^-$ become geared below $T$$_N$. Since the periodic motions involve concerted motion of net charges, the associated magnetic moments of the NO$_3$$^-$ ions indirectly establish an antiferromagnetic structure below $T$$_N$. Our findings identify a previously unidentified type of molecular interaction which may be exploited to further increase the enthalpy of the widely-popular hydrated salts employed as energy storage devices., Comment: 13 pages (single column, 1.5 spaced), 5 figures

Data-driven soft sensors (DDSS) have become mainstream methods for predicting key performance indicators in process industries. However, DDSS development requires complex and costly customized designs tailored to various tasks during the modeling process. Moreover, DDSS are constrained to a single structured data modality, limiting their ability to incorporate additional contextual knowledge. Furthermore, DDSSs' limited representation learning leads to weak predictive performance with scarce data. To address these challenges, we propose a general framework named LLM-TKESS (large language model for text-based knowledge-embedded soft sensing), harnessing the powerful general problem-solving capabilities, cross-modal knowledge transfer abilities, and few-shot capabilities of LLM for enhanced soft sensing modeling. Specifically, an auxiliary variable series encoder (AVS Encoder) is proposed to unleash LLM's potential for capturing temporal relationships within series and spatial semantic relationships among auxiliary variables. Then, we propose a two-stage fine-tuning alignment strategy: in the first stage, employing parameter-efficient fine-tuning through autoregressive training adjusts LLM to rapidly accommodate process variable data, resulting in a soft sensing foundation model (SSFM). Subsequently, by training adapters, we adapt the SSFM to various downstream tasks without modifying its architecture. Then, we propose two text-based knowledge-embedded soft sensors, integrating new natural language modalities to overcome the limitations of pure structured data models. Furthermore, benefiting from LLM's pre-existing world knowledge, our model demonstrates outstanding predictive capabilities in small sample conditions. Using the thermal deformation of air preheater rotor as a case study, we validate through extensive experiments that LLM-TKESS exhibits outstanding performance.

We present the Metadetection weak lensing galaxy shape catalogue from the six-year Dark Energy Survey (DES Y6) imaging data. This dataset is the final release from DES, spanning 4422 deg$^2$ of the southern sky. We describe how the catalogue was constructed, including the two new major processing steps, cell-based image coaddition and shear measurements with Metadetection. The DES Y6 Metadetection weak lensing shape catalogue consists of 151,922,791 galaxies detected over riz bands, with an effective number density of $n_{\rm eff}$ =8.22 galaxies per arcmin$^2$ and shape noise of $\sigma_e$ = 0.29. We carry out a suite of validation tests on the catalogue, including testing for PSF leakage, testing for the impact of PSF modeling errors, and testing the correlation of the shear measurements with galaxy, PSF, and survey properties. In addition to demonstrating that our catalogue is robust for weak lensing science, we use the DES Y6 image simulation suite (Mau, Becker et al. 2025) to estimate the overall multiplicative shear bias of our shear measurement pipeline. We find no detectable multiplicative bias at the roughly half-percent level, with m = (3.4 $\pm$ 6.1) x $10^{-3}$, at 3$\sigma$ uncertainty. This is the first time both cell-based coaddition and Metadetection algorithms are applied to observational data, paving the way to the Stage-IV weak lensing surveys., Comment: 30 pages, 22 figures

The surface tension coefficient is a key parameter in fluid mechanics. The conventional method to measure it is to determine the critical surface tension that causes the rupture of a liquid film. However, this method has a large error because the surface tension coefficient is very sensitive to environmental factors. In this paper, we propose a new method based on equal thickness interference to study the mathematical model of the stationary liquid surface shape and measure it by the interference technique. This method enables the accurate measurement of the liquid surface tension coefficient. The experimental results demonstrate that this method has the advantages of high accuracy, low cost, and simple instrumentation. The maximum measurement error is less than 3.1\%. Moreover, this method can be applied to measure the liquid-liquid interfacial tension, which has a good application prospect.

Recent advancements in multimodal models have shown a strong ability in visual perception, reasoning abilities, and vision-language understanding. However, studies on visual matching ability are missing, where finding the visual correspondence of objects is essential in vision research. Our research reveals that the matching capabilities in recent multimodal LLMs (MLLMs) still exhibit systematic shortcomings, even with current strong MLLMs models, GPT-4o. In particular, we construct a Multimodal Visual Matching (MMVM) benchmark to fairly benchmark over 30 different MLLMs. The MMVM benchmark is built from 15 open-source datasets and Internet videos with manual annotation. We categorize the data samples of MMVM benchmark into eight aspects based on the required cues and capabilities to more comprehensively evaluate and analyze current MLLMs. In addition, we have designed an automatic annotation pipeline to generate the MMVM SFT dataset, including 220K visual matching data with reasoning annotation. Finally, we present CoLVA, a novel contrastive MLLM with two novel technical designs: fine-grained vision expert with object-level contrastive learning and instruction augmentation strategy. CoLVA achieves 51.06\% overall accuracy (OA) on the MMVM benchmark, surpassing GPT-4o and baseline by 8.41\% and 23.58\% OA, respectively. The results show the effectiveness of our MMVM SFT dataset and our novel technical designs. Code, benchmark, dataset, and models are available at https://github.com/zhouyiks/CoLVA., Comment: project page: https://zhouyiks.github.io/projects/CoLVA/

The correlation of topology and disorder has attracted great intention due to appropriate disorder could induce the phase transition between trivial and nontrivial topological states. While it is widely recognized that strong disorder can produce rich phase diagrams in topological nontrivial states, moderate disorder has been proposed to induce transitions into topologically nontrivial phases counter-intuitively, leading to the concept of topological Anderson insulators. This phenomenon has been theoretically explored and simulated in various systems, yet experimental realization in solid state systems has remained elusive due to challenges in controlling disorder. Here, we report the experimental observation of Chern insulator state signed by the coexistence of quantized Hall plateau and zero longitudinal resistance in monolayer MnBi$_4$Te$_7$ Hall bar device, which originally hosts a trivial insulating state with Chern number $C$ = 0 in clean limit. We demonstrate that the observed trivial to nontrivial transition in this monolayer device can be attributed to disorder, evidenced by universal conductance fluctuations. Our findings substantiate the existence of a long-sought topological Anderson Chern insulator in real materials, a unique variant of the topological Anderson insulator characterized by broken time-reversal-symmetry., Comment: 45 pages, 4 main figures, 5 extended data figures, 5 supplementary figures

Long-form video understanding with Large Vision Language Models is challenged by the need to analyze temporally dispersed yet spatially concentrated key moments within limited context windows. In this work, we introduce VideoMindPalace, a new framework inspired by the "Mind Palace", which organizes critical video moments into a topologically structured semantic graph. VideoMindPalace organizes key information through (i) hand-object tracking and interaction, (ii) clustered activity zones representing specific areas of recurring activities, and (iii) environment layout mapping, allowing natural language parsing by LLMs to provide grounded insights on spatio-temporal and 3D context. In addition, we propose the Video MindPalace Benchmark (VMB), to assess human-like reasoning, including spatial localization, temporal reasoning, and layout-aware sequential understanding. Evaluated on VMB and established video QA datasets, including EgoSchema, NExT-QA, IntentQA, and the Active Memories Benchmark, VideoMindPalace demonstrates notable gains in spatio-temporal coherence and human-aligned reasoning, advancing long-form video analysis capabilities in VLMs.

This work presents Sa2VA, the first unified model for dense grounded understanding of both images and videos. Unlike existing multi-modal large language models, which are often limited to specific modalities and tasks, Sa2VA supports a wide range of image and video tasks, including referring segmentation and conversation, with minimal one-shot instruction tuning. Sa2VA combines SAM-2, a foundation video segmentation model, with LLaVA, an advanced vision-language model, and unifies text, image, and video into a shared LLM token space. Using the LLM, Sa2VA generates instruction tokens that guide SAM-2 in producing precise masks, enabling a grounded, multi-modal understanding of both static and dynamic visual content. Additionally, we introduce Ref-SAV, an auto-labeled dataset containing over 72k object expressions in complex video scenes, designed to boost model performance. We also manually validate 2k video objects in the Ref-SAV datasets to benchmark referring video object segmentation in complex environments. Experiments show that Sa2VA achieves state-of-the-art across multiple tasks, particularly in referring video object segmentation, highlighting its potential for complex real-world applications., Comment: Project page: https://lxtgh.github.io/project/sa2va

Despite the significant improvements achieved by large language models (LLMs) in English reasoning tasks, these models continue to struggle with multilingual reasoning. Recent studies leverage a full-parameter and two-stage training paradigm to teach models to first understand non-English questions and then reason. However, this method suffers from both substantial computational resource computing and catastrophic forgetting. The fundamental cause is that, with the primary goal of enhancing multilingual comprehension, an excessive number of irrelevant layers and parameters are tuned during the first stage. Given our findings that the representation learning of languages is merely conducted in lower-level layers, we propose an efficient multilingual reasoning alignment approach that precisely identifies and fine-tunes the layers responsible for handling multilingualism. Experimental results show that our method, SLAM, only tunes 6 layers' feed-forward sub-layers including 6.5-8% of all parameters within 7B and 13B LLMs, achieving superior average performance than all strong baselines across 10 languages. Meanwhile, SLAM only involves one training stage, reducing training time by 4.1-11.9 compared to the two-stage method., Comment: Accepted by COLING 2025 (Oral)

Ultra-fast electronic phenomena originating from finite temperature, such as nonlinear optical excitation, can be simulated with high fidelity via real-time time dependent density functional theory (rt-TDDFT) calculations with hybrid functional. However, previous rt-TDDFT simulations of real materials using the optimal gauge--known as the parallel transport gauge--have been limited to low-temperature systems with band gaps. In this paper, we introduce the parallel transport-implicit midpoint (PT-IM) method, which significantly accelerates finite-temperature rt-TDDFT calculations of real materials with hybrid function. We first implement PT-IM with hybrid functional in our plane wave code PWDFT, and optimized it on both GPU and ARM platforms to build a solid baseline code. Next, we propose a diagonalization method to reduce computation and communication complexity, and then, we employ adaptively compressed exchange (ACE) method to reduce the frequency of the most expensive Fock exchange operator. Finally, we adopt the ring\_based method and the shared memory mechanism to overlap computation and communication and alleviate memory consumption respectively. Numerical results show that our optimized code can reach 3072 atoms for rt-TDDFT simulation with hybrid functional at finite temperature on 192 computing nodes, the time-to-solution for one time step is 429.3s, which is 41.4 times faster compared to the baseline.

Fluid antennas present a relatively new idea for harnessing the fading and interference issues in multiple user wireless systems, such as 6G. Here, we systematically compare their unique radiation beam forming mechanism to the existing multiple-antenna systems in a wireless system. Subsequently, a unified mathematical model for fluid antennas is deduced based on the eigenmode theory. As mathematically derived from the multimode resonant theory, the spectral expansion model of any antennas which occupy variable spaces and have changeable feeding schemes can be generalized as fluid antennas. Non-liquid and liquid fluid antenna examples are presented, simulated and discussed. The symmetry or modal parity of eigenmodes is explored as an additional degree of freedom to design the fluid antennas for future wireless systems. As conceptually deduced and illustrated, the multi-dimensional and continuously adaptive ability of eigenmodes can be considered as the most fundamental intrinsic characteristic of the fluid antenna systems. It opens an uncharted area in the developments of intelligent antennas (IAs), which brings more flexibility to on-demand antenna beam null manipulating techniques for future wireless applications.