Search

Existing image steganography methods face fundamental limitations in hiding capacity (typically $1\sim7$ images) due to severe information interference and uncoordinated capacity-distortion trade-off. We propose SMILENet, a novel synergistic framework that achieves 25 image hiding through three key innovations: (i) A synergistic network architecture coordinates reversible and non-reversible operations to efficiently exploit information redundancy in both secret and cover images. The reversible Invertible Cover-Driven Mosaic (ICDM) module and Invertible Mosaic Secret Embedding (IMSE) module establish cover-guided mosaic transformations and representation embedding with mathematically guaranteed invertibility for distortion-free embedding. The non-reversible Secret Information Selection (SIS) module and Secret Detail Enhancement (SDE) module implement learnable feature modulation for critical information selection and enhancement. (ii) A unified training strategy that coordinates complementary modules to achieve 3.0x higher capacity than existing methods with superior visual quality. (iii) Last but not least, we introduce a new metric to model Capacity-Distortion Trade-off for evaluating the image steganography algorithms that jointly considers hiding capacity and distortion, and provides a unified evaluation approach for accessing results with different number of secret image. Extensive experiments on DIV2K, Paris StreetView and ImageNet1K show that SMILENet outperforms state-of-the-art methods in terms of hiding capacity, recovery quality as well as security against steganalysis methods.

The emergence of 3D Gaussian Splatting (3D-GS) has significantly advanced 3D reconstruction by providing high fidelity and fast training speeds across various scenarios. While recent efforts have mainly focused on improving model structures to compress data volume or reduce artifacts during zoom-in and zoom-out operations, they often overlook an underlying issue: training sampling deficiency. In zoomed-in views, Gaussian primitives can appear unregulated and distorted due to their dilation limitations and the insufficient availability of scale-specific training samples. Consequently, incorporating pseudo-details that ensure the completeness and alignment of the scene becomes essential. In this paper, we introduce a new training method that integrates diffusion models and multi-scale training using pseudo-ground-truth data. This approach not only notably mitigates the dilation and zoomed-in artifacts but also enriches reconstructed scenes with precise details out of existing scenarios. Our method achieves state-of-the-art performance across various benchmarks and extends the capabilities of 3D reconstruction beyond training datasets.

Single Image Reflection Removal (SIRR) is a canonical blind source separation problem and refers to the issue of separating a reflection-contaminated image into a transmission and a reflection image. The core challenge lies in minimizing the commonalities among different sources. Existing deep learning approaches either neglect the significance of feature interactions or rely on heuristically designed architectures. In this paper, we propose a novel Deep Exclusion unfolding Network (DExNet), a lightweight, interpretable, and effective network architecture for SIRR. DExNet is principally constructed by unfolding and parameterizing a simple iterative Sparse and Auxiliary Feature Update (i-SAFU) algorithm, which is specifically designed to solve a new model-based SIRR optimization formulation incorporating a general exclusion prior. This general exclusion prior enables the unfolded SAFU module to inherently identify and penalize commonalities between the transmission and reflection features, ensuring more accurate separation. The principled design of DExNet not only enhances its interpretability but also significantly improves its performance. Comprehensive experiments on four benchmark datasets demonstrate that DExNet achieves state-of-the-art visual and quantitative results while utilizing only approximately 8\% of the parameters required by leading methods.

Graph Neural Networks (GNNs) are powerful tools for processing graph-structured data, increasingly used for large-scale real-world graphs via sampling-based inference methods. However, inherent characteristics of neighbor sampling lead to redundant data loading during GNN inference, compounded by inefficient data transfers between host and GPU memory, resulting in slow inference and low resource utilization. Existing methods to accelerate GNN inference face several challenges: (1) low practical GPU memory utilization, (2) overlooking adjacency matrix locality, and (3) long preprocessing time. To address these challenges, we introduce DCI, an efficient workload-aware dual-cache allocation system for GNN inference acceleration. DCI allocates cache capacities for both node features and adjacency matrices based on workload patterns during the pre-sampling phase, leveraging a lightweight cache-filling algorithm to optimize data loading efficiency. Experimental results demonstrate that DCI accelerates sampling and node feature loading, achieving end-to-end inference speedups of 1.18$\times$ to 11.26$\times$ compared to DGL, and 1.14$\times$ to 13.68$\times$ over RAIN, while reducing preprocessing time by 52.8\% to 98.7\%. Additionally, DCI outperforms state-of-the-art single-cache inference systems by achieving speedup of 1.08$\times$ to 1.32$\times$. We also compared DCI with DUCATI's dual-cache population strategy. Our lightweight population algorithm allows DCI to achieve nearly the same inference speed while keeping preprocessing time to less than 20\% of that required by DUCATI.

Nanoscopic heterogeneities in composition and structure are quintessential for the properties of electrocatalyst materials. Here, we present a semiclassical model to study the electrochemical properties of supported electrocatalyst nanoparticles (NP). The model captures the correlated electronic and ionic equilibration across NP, support, and electrolyte. It reveals peculiar trends in surface charging of the supported NP, validated by comparison with first-principles calculations. Support-induced perturbations in electronic and ionic charge densities at the NP's active surface manifest as distinct potentials of zero local electronic and ionic charges that could differ by more than 0.5 V in the studied system.

The increasing focus on predicting renewable energy production aligns with advancements in deep learning (DL). The inherent variability of renewable sources and the complexity of prediction methods require robust approaches, such as DL models, in the renewable energy sector. DL models are preferred over traditional machine learning (ML) because they capture complex, nonlinear relationships in renewable energy datasets. This study examines key factors influencing DL technique accuracy, including sampling and hyperparameter optimization, by comparing various methods and training and test ratios within a DL framework. Seven machine learning methods, LSTM, Stacked LSTM, CNN, CNN-LSTM, DNN, Time-Distributed MLP (TD-MLP), and Autoencoder (AE), are evaluated using a dataset combining weather and photovoltaic power output data from 12 locations. Regularization techniques such as early stopping, neuron dropout, L1 and L2 regularization are applied to address overfitting. The results demonstrate that the combination of early stopping, dropout, and L1 regularization provides the best performance to reduce overfitting in the CNN and TD-MLP models with larger training set, while the combination of early stopping, dropout, and L2 regularization is the most effective to reduce the overfitting in CNN-LSTM and AE models with smaller training set., Comment: 11 pages, 2 figures and 6 tables

The determination of the equation of state (EOS) of a neutron star (NS) and its maximum mass is very important for understanding the formation and properties of NSs under extreme conditions, but they remain open questions. Short-duration gamma-ray bursts (GRBs) are believed to originate from the merger of binary NSs or giant flares (GFs) of soft gamma repeaters (SGRs). Recently, the high-frequency quasi-periodic oscillations (QPOs) have been claimed to be identified from two short GRBs (GRB 931101B and GRB 910711). In this paper, we propose that the observed high-frequency QPOs in these two short GRBs result from torsional oscillations in the GFs of SGRs associated with cold NSs, or from radial oscillations of hypermassive NSs as the hot remnants of binary NS mergers, and then to constrain the EOS of NSs. For torsional oscillations, the six selected EOSs (TM1, NL3, APR, SLy4, DDME2, and GM1) of NSs suitable for the zero-temperature condition exhibit significant overlap in mass ranges, suggesting that we cannot constrain the EOS of NSs. For radial oscillations, the six selected EOSs (IUF, TM1, TMA, FSG, BHBLp, and NL3) of NSs suitable for the high-temperature condition cannot be ruled out when redshift is considered. However, it is found that the EOS can only be constrained if the redshift and temperature of the remnant can be measured., Comment: 16 pages, 2 Tables, 10 Figures, accepted for publication in ApJ, and matched with the published verison

Low-altitude mobility, exemplified by unmanned aerial vehicles (UAVs), has introduced transformative advancements across various domains, like transportation, logistics, and agriculture. Leveraging flexible perspectives and rapid maneuverability, UAVs extend traditional systems' perception and action capabilities, garnering widespread attention from academia and industry. However, current UAV operations primarily depend on human control, with only limited autonomy in simple scenarios, and lack the intelligence and adaptability needed for more complex environments and tasks. The emergence of large language models (LLMs) demonstrates remarkable problem-solving and generalization capabilities, offering a promising pathway for advancing UAV intelligence. This paper explores the integration of LLMs and UAVs, beginning with an overview of UAV systems' fundamental components and functionalities, followed by an overview of the state-of-the-art in LLM technology. Subsequently, it systematically highlights the multimodal data resources available for UAVs, which provide critical support for training and evaluation. Furthermore, it categorizes and analyzes key tasks and application scenarios where UAVs and LLMs converge. Finally, a reference roadmap towards agentic UAVs is proposed, aiming to enable UAVs to achieve agentic intelligence through autonomous perception, memory, reasoning, and tool utilization. Related resources are available at https://github.com/Hub-Tian/UAVs_Meet_LLMs.

The burgeoning field of foundation models necessitates advanced data processing mechanisms capable of harnessing vast valuable data with varied types utilized by these models. Nevertheless, the current landscape presents unique challenges that traditional data processing frameworks cannot handle effectively, especially with multimodal intricacies. In response, we present Data-Juicer 2.0, a new system offering fruitful data processing capabilities backed by over a hundred operators spanning various modalities like text, image, audio, and video. With seamless compatibility and dedicated optimization to popular dataset hubs like Hugging Face and computing engines like Ray, Data-Juicer 2.0 enhances its predecessor in both usability, efficiency, and programmability. It features an easily accessible user interface layer that supports decoupled Python interactions, RESTful APIs, and conversational commands. Alongside this, it contains a core runtime layer optimized for adaptive execution and management across different dataset scales, processing demands, and computational environments, while shielding unnecessary system details. Extensive empirical evaluations demonstrate Data-Juicer 2.0's remarkable performance and scalability, highlighting its capability to efficiently process tens of billions of data samples with tens of thousands of CPU cores. The system is publicly available, actively maintained, and broadly adopted in diverse research endeavors, practical applications, and real-world products such as Alibaba Cloud PAI., Comment: 16 pages, 9 figures, 3 tables

We present simulation results examining the presence and behavior of standing shocks in zero-energy low angular momentum advective accretion flows and explore their (in)stabilities properties taking into account various specific angular momentum, $\lambda_0$. Within the range $10-50R_g$ (where $R_g$ denotes the Schwarzschild radius), shocks are discernible for $\lambda_0\geq 1.75$. In the special relativistic hydrodynamic (RHD) simulation when $\lambda_0 = 1.80$, we find the merger of two shocks resulted in a dramatic increase in luminosity. We present the impact of external and internal flow collisions from the funnel region on luminosity. Notably, oscillatory behavior characterizes shocks within $1.70 \leq \lambda_0 \leq 1.80$. Using free-free emission as a proxy for analysis, we shows that the luminosity oscillations between frequencies of $0.1-10$ Hz for $\lambda_0$ range $1.7 \leq \lambda_0 \leq 1.80$. These findings offer insights into quasi-periodic oscillations emissions from certain black hole X-ray binaries, exemplified by GX 339-4. Furthermore, for the supermassive black hole at the Milky Way's center, Sgr A*, oscillation frequencies between $10^{-6}$ and $10^{-5}$ Hz were observed. This frequency range, translating to one cycle every few days, aligns with observational data from the X-ray telescopes such as Chandra, Swift, and XMM-Newton., Comment: 22 pages, 13 figures, Accepted for publication in the Research in Astronomy and Astrophysics (RAA) journal

High silicon steel with 6.5% silicon content is the best because of its excellent magnetic properties, such as high saturation magnetization, high resistivity, low iron loss and near zero magnetostriction. High silicon steel can greatly save energy, and reduce the weight and size of electrical appliances. This has a very important application prospect for energy and aerospace industry. The high brittleness of high silicon steel makes its production and processing very difficult. For more than 30 years, many steel companies and research institutions around the world have adopted various technical means to study the industrialization of high silicon steel, but they have not been successful . JFE-NKK steel company in Japan has realized the small batch production of high silicon steel by using SiCl4-CVD technology. However, due to the complex process, corrosion and pollution, high cost, its production scale is greatly limited. So far, large-scale production of high silicon steel is still a major challenge in the world. This paper will introduce the experimental results of successfully preparing high silicon steel by Double Glow Plasma Surface Metallurgy Technology. The process is simple and easy without any corrosion or pollution, which may provide a new way for the world to achieve large-scale production of high silicon steel. The large-scale production and wide application of high silicon steel is likely to change the pattern of the world's energy and electric power industry, save a lot of energy for mankind, and create huge economic benefits.

Specializing LLMs in various domain-specific tasks has emerged as a critical step towards achieving high performance. However, the construction and annotation of datasets in specific domains are always very costly. Apart from using superior and expensive closed-source LLM APIs to construct datasets, some open-source models have become strong enough to handle dataset construction in many scenarios. Thus, we present a family of data augmentation models designed to significantly improve the efficiency for model fine-tuning. These models, trained based on sufficiently small LLMs, support key functionalities with low inference costs: instruction expansion, instruction refinement, and instruction-response pair expansion. To fulfill this goal, we first construct an automatic data collection system with seed datasets generated from both public repositories and our in-house datasets. This system leverages powerful LLMs to expand, refine and re-write the instructions and responses, incorporating quality assessment techniques. Following this, we introduce the training process of our models, which effectively distills task-solving and text synthesis abilities from teacher LLMs. Finally, we demonstrate how we integrate these functionalities into a machine learning platform to support low-cost LLM fine-tuning from both dataset preparation and training perspectives for users. Experiments and an application study prove the effectiveness of our approach., Comment: coling 2025 industry track

Optically-addressable spin qubits form the foundation of a new generation of emerging nanoscale sensors. The engineering of these sensors has mainly focused on solid-state systems such as the nitrogen-vacancy (NV) center in diamond. However, NVs are restricted in their ability to interface with biomolecules due to their bulky diamond host. Meanwhile, fluorescent proteins have become the gold standard in bioimaging, as they are genetically encodable and easily integrated with biomolecules. While fluorescent proteins have been suggested to possess a metastable triplet state, they have not been investigated as qubit sensors. Here, we realize an optically-addressable spin qubit in the Enhanced Yellow Fluorescent Protein (EYFP) enabled by a novel spin-readout technique. A near-infrared laser pulse allows for triggered readout of the triplet state with up to 44% spin contrast. Using coherent microwave control of the EYFP spin at liquid-nitrogen temperatures, we measure a spin-lattice relaxation time of $(141 \pm 5)\, \mathrm{\mu s}$, a $(16 \pm 2)\, \mathrm{\mu s}$ coherence time under Carr-Purcell-Meiboom-Gill (CPMG) decoupling, and a predicted oscillating (AC) magnetic field sensitivity with an upper bound of $183 \, \mathrm{fT}\, \mathrm{mol}^{1/2}\, \mathrm{Hz}^{-1/2}$. We express the qubit in mammalian cells, maintaining contrast and coherent control despite the complex intracellular environment. Finally, we demonstrate optically-detected magnetic resonance at room temperature in aqueous solution with contrast up to 3%, and measure a static (DC) field sensitivity with an upper bound of $93 \, \mathrm{pT}\, \mathrm{mol}^{1/2}\, \mathrm{Hz}^{-1/2}$. Our results establish fluorescent proteins as a powerful new qubit sensor platform and pave the way for applications in the life sciences that are out of reach for solid-state technologies.

A classical coulombic correlation functional in one-loop (1L) and local-density-approximation (LDA) is derived for electrolyte solutions, starting from a first-principles many-body partition function. The 1L-LDA functional captures correlations between electrolyte ions and solvent dipoles, such as screening and solvation, that are ignored by conventional mean-field theories. This 1L-LDA functional introduces two parameters that can be tuned to the experimental dielectric permittivity and activity coefficients in the bulk electrolyte solution. The capabilities of the 1L-LDA functional for the description of metal-electrolyte interfaces are demonstrated by embedding the functional into a combined quantum-classical model. Here, the 1L-LDA functional leads to a more pronounced double-peak structure of the interfacial capacitance with higher peaks and shorter peak-to-peak distance, significantly improving the agreement with experimental data and showing that electrolyte correlation effects exert a vital impact on the capacitive response.

This article presents the design of an open-API-based explainable AI (XAI) service to provide feature contribution explanations for cloud AI services. Cloud AI services are widely used to develop domain-specific applications with precise learning metrics. However, the underlying cloud AI services remain opaque on how the model produces the prediction. We argue that XAI operations are accessible as open APIs to enable the consolidation of the XAI operations into the cloud AI services assessment. We propose a design using a microservice architecture that offers feature contribution explanations for cloud AI services without unfolding the network structure of the cloud models. We can also utilize this architecture to evaluate the model performance and XAI consistency metrics showing cloud AI services trustworthiness. We collect provenance data from operational pipelines to enable reproducibility within the XAI service. Furthermore, we present the discovery scenarios for the experimental tests regarding model performance and XAI consistency metrics for the leading cloud vision AI services. The results confirm that the architecture, based on open APIs, is cloud-agnostic. Additionally, data augmentations result in measurable improvements in XAI consistency metrics for cloud AI services., Comment: Published in: IEEE Transactions on Cloud Computing ( Volume: 12, Issue: 2, April-June 2024)

To meet the growing demand for accurate, rapid, and cost-effective at-home clinical testing, we developed a radical-mediated enzyme assay (REEA) integrated with a paper fluidic system and electrically read by a handheld field-effect transistor (FET) device. The REEA utilizes horseradish peroxidase (HRP) to catalyze the conversion of aromatic substrates into radical forms, producing protons detected by an ion-sensitive FET for biomarker quantification. Through screening 14 phenolic compounds, halogenated phenols emerged as optimal substrates for the REEA. Encased in an affordable cartridge ($0.55 per test), the system achieved a detection limit of 146 fg/mL for estradiol (E2), with a coefficient of variation (CV) below 9.2% in E2-spiked samples and an r2 of 0.963 across a measuring range of 19 to 4,551 pg/mL in clinical plasma samples, providing results in under 10 minutes. This adaptable system not only promises to offer a fast and reliable platform, but also holds significant potential for expansion to a wide array of biomarkers, paving the way for broader clinical and home-based applications.

In applications, an anticipated situation is where the system of interest has never been encountered before and sparse observations can be made only once. Can the dynamics be faithfully reconstructed from the limited observations without any training data? This problem defies any known traditional methods of nonlinear time-series analysis as well as existing machine-learning methods that typically require extensive data from the target system for training. We address this challenge by developing a hybrid transformer and reservoir-computing machine-learning scheme. The key idea is that, for a complex and nonlinear target system, the training of the transformer can be conducted not using any data from the target system, but with essentially unlimited synthetic data from known chaotic systems. The trained transformer is then tested with the sparse data from the target system. The output of the transformer is further fed into a reservoir computer for predicting the long-term dynamics or the attractor of the target system. The power of the proposed hybrid machine-learning framework is demonstrated using a large number of prototypical nonlinear dynamical systems, with high reconstruction accuracy even when the available data is only 20% of that required to faithfully represent the dynamical behavior of the underlying system. The framework provides a paradigm of reconstructing complex and nonlinear dynamics in the extreme situation where training data does not exist and the observations are random and sparse., Comment: 31 pages, 21 figures

Accurate interpretation and visualization of human instructions are crucial for text-to-image (T2I) synthesis. However, current models struggle to capture semantic variations from word order changes, and existing evaluations, relying on indirect metrics like text-image similarity, fail to reliably assess these challenges. This often obscures poor performance on complex or uncommon linguistic patterns by the focus on frequent word combinations. To address these deficiencies, we propose a novel metric called SemVarEffect and a benchmark named SemVarBench, designed to evaluate the causality between semantic variations in inputs and outputs in T2I synthesis. Semantic variations are achieved through two types of linguistic permutations, while avoiding easily predictable literal variations. Experiments reveal that the CogView-3-Plus and Ideogram 2 performed the best, achieving a score of 0.2/1. Semantic variations in object relations are less understood than attributes, scoring 0.07/1 compared to 0.17-0.19/1. We found that cross-modal alignment in UNet or Transformers plays a crucial role in handling semantic variations, a factor previously overlooked by a focus on textual encoders. Our work establishes an effective evaluation framework that advances the T2I synthesis community's exploration of human instruction understanding. Our benchmark and code are available at https://github.com/zhuxiangru/SemVarBench ., Comment: The only change in the current version update is the replacement of the template with a more precise one

Contrastive Language-Image Pre-training (CLIP) has been widely studied and applied in numerous applications. However, the emphasis on brief summary texts during pre-training prevents CLIP from understanding long descriptions. This issue is particularly acute regarding videos given that videos often contain abundant detailed contents. In this paper, we propose the VideoCLIP-XL (eXtra Length) model, which aims to unleash the long-description understanding capability of video CLIP models. Firstly, we establish an automatic data collection system and gather a large-scale VILD pre-training dataset with VIdeo and Long-Description pairs. Then, we propose Text-similarity-guided Primary Component Matching (TPCM) to better learn the distribution of feature space while expanding the long description capability. We also introduce two new tasks namely Detail-aware Description Ranking (DDR) and Hallucination-aware Description Ranking (HDR) for further understanding improvement. Finally, we construct a Long Video Description Ranking (LVDR) benchmark for evaluating the long-description capability more comprehensively. Extensive experimental results on widely-used text-video retrieval benchmarks with both short and long descriptions and our LVDR benchmark can fully demonstrate the effectiveness of our method., Comment: EMNLP 2024 Main conference

Text-rich document understanding (TDU) refers to analyzing and comprehending documents containing substantial textual content. With the rapid evolution of large language models (LLMs), they have been widely leveraged for TDU due to their remarkable versatility and generalization. In this paper, we introduce DocLayLLM, an efficient and effective multi-modal extension of LLMs specifically designed for TDU. By integrating visual patch tokens and 2D positional tokens into LLMs and encoding the document content using the LLMs themselves, we fully take advantage of the document comprehension capability of LLMs and enhance their perception of OCR information. We have also deeply considered the role of the chain-of-thought (CoT) and innovatively proposed the techniques of CoT Pre-training and CoT Annealing. Our DocLayLLM can achieve remarkable performances with lightweight training settings, showcasing its efficiency and effectiveness. Experimental results demonstrate that our DocLayLLM surpasses existing OCR-dependent methods and also outperforms OCR-free competitors.

Non-blind rotary motion deblurring (RMD) aims to recover the latent clear image from a rotary motion blurred (RMB) image. The rotation center is a crucial input parameter in non-blind RMD methods. Existing methods directly estimate the rotation center from the RMB image. However they always suffer significant errors, and the performance of RMD is limited. For the assembled imaging systems, the position of the rotation center remains fixed. Leveraging this prior knowledge, we propose a geometric-based method for rotation center identification and analyze its error range. Furthermore, we construct a RMB imaging system. The experiment demonstrates that our method achieves less than 1-pixel error along a single axis (x-axis or y-axis). We utilize the constructed imaging system to capture real RMB images, and experimental results show that our method can help existing RMD approaches yield better RMD images.

Inferring future activity information based on observed activity data is a crucial step to improve the accuracy of early activity prediction. Traditional methods based on generative adversarial networks(GAN) or joint learning frameworks can achieve good prediction accuracy under low observation ratios, but they usually have high computational costs. In view of this, this paper proposes a spatio-temporal encoding and decoding-based method for future human activity skeleton synthesis. Firstly, algorithms such as time control, discrete cosine transform, and low-pass filtering are used to cut or pad the skeleton sequences. Secondly, the encoder and decoder are responsible for extracting intermediate semantic encoding from observed skeleton sequences and inferring future sequences from the intermediate semantic encoding, respectively. Finally, joint displacement error, velocity error, and acceleration error, three higher-order kinematic features, are used as key components of the loss function to optimize model parameters. Experimental results show that the proposed future skeleton synthesis algorithm performs better than some existing algorithms. It generates skeleton sequences with smaller errors and fewer model parameters, effectively providing future information for early activity prediction.

Retrieval augmented generation (RAG) exhibits outstanding performance in promoting the knowledge capabilities of large language models (LLMs) with retrieved documents related to user queries. However, RAG only focuses on improving the response quality of LLMs via enhancing queries indiscriminately with retrieved information, paying little attention to what type of knowledge LLMs really need to answer original queries more accurately. In this paper, we suggest that long-tail knowledge is crucial for RAG as LLMs have already remembered common world knowledge during large-scale pre-training. Based on our observation, we propose a simple but effective long-tail knowledge detection method for LLMs. Specifically, the novel Generative Expected Calibration Error (GECE) metric is derived to measure the ``long-tailness'' of knowledge based on both statistics and semantics. Hence, we retrieve relevant documents and infuse them into the model for patching knowledge loopholes only when the input query relates to long-tail knowledge. Experiments show that, compared to existing RAG pipelines, our method achieves over 4x speedup in average inference time and consistent performance improvement in downstream tasks.

The COVID-19 pandemic has compelled multinational corporations to diversify their global supply chain risk and to relocate their factories to Southeast Asian countries beyond China. Such recent phenomena provide a good opportunity to understand the factors that influenced offshore decisions in the last two decades. We propose a new conceptual framework based on econometric approaches to examine the relationships between these factors. Firstly, the Vector Auto Regression (VAR) for multi-way cointegration analysis by a Johansen test as well as the embedding Granger causality analysis to examine offshore decisions--innovation, technology readiness, infrastructure, foreign direct investment (FDI), and intermediate imports. Secondly, a Quantile Vector Autoregressive (QVAR) model is used to assess the dynamic connectedness among Southeast Asian countries based on the offshore factors. This study explores a system-wide experiment to evaluate the spillover effects of offshore decisions. It reports a comprehensive analysis using time-series data collected from the World Bank. The results of the cointegration, causality, and dynamic connectedness analyses show that a subset of Southeast Asian countries have spillover effects on each other. These countries present a multi-way cointegration and dynamic connectedness relationship. The study contributes to policymaking by providing a data-driven innovative approach through a new conceptual framework., Comment: 30 pages

Robust and imperceptible adversarial video attack is challenging due to the spatial and temporal characteristics of videos. The existing video adversarial attack methods mainly take a gradient-based approach and generate adversarial videos with noticeable perturbations. In this paper, we propose a novel Sparse Adversarial Video Attack via Spatio-Temporal Invertible Neural Networks (SVASTIN) to generate adversarial videos through spatio-temporal feature space information exchanging. It consists of a Guided Target Video Learning (GTVL) module to balance the perturbation budget and optimization speed and a Spatio-Temporal Invertible Neural Network (STIN) module to perform spatio-temporal feature space information exchanging between a source video and the target feature tensor learned by GTVL module. Extensive experiments on UCF-101 and Kinetics-400 demonstrate that our proposed SVASTIN can generate adversarial examples with higher imperceptibility than the state-of-the-art methods with the higher fooling rate. Code is available at \href{https://github.com/Brittany-Chen/SVASTIN}{https://github.com/Brittany-Chen/SVASTIN}.

Recently, while large language models (LLMs) have demonstrated impressive results, they still suffer from hallucination, i.e., the generation of false information. Model editing is the task of fixing factual mistakes in LLMs; yet, most previous works treat it as a one-time task, paying little attention to ever-emerging mistakes generated by LLMs. We address the task of sequential model editing (SME) that aims to rectify mistakes continuously. A Dynamic Auxiliary Fusion Network (DAFNet) is designed to enhance the semantic interaction among the factual knowledge within the entire sequence, preventing catastrophic forgetting during the editing process of multiple knowledge triples. Specifically, (1) for semantic fusion within a relation triple, we aggregate the intra-editing attention flow into auto-regressive self-attention with token-level granularity in LLMs. We further leverage multi-layer diagonal inter-editing attention flow to update the weighted representations of the entire sequence-level granularity. (2) Considering that auxiliary parameters are required to store the knowledge for sequential editing, we construct a new dataset named \textbf{DAFSet}, fulfilling recent, popular, long-tail and robust properties to enhance the generality of sequential editing. Experiments show DAFNet significantly outperforms strong baselines in single-turn and sequential editing. The usage of DAFSet also consistently improves the performance of other auxiliary network-based methods in various scenarios, Comment: ACL2024 findings

This paper presents EasyAnimate, an advanced method for video generation that leverages the power of transformer architecture for high-performance outcomes. We have expanded the DiT framework originally designed for 2D image synthesis to accommodate the complexities of 3D video generation by incorporating a motion module block. It is used to capture temporal dynamics, thereby ensuring the production of consistent frames and seamless motion transitions. The motion module can be adapted to various DiT baseline methods to generate video with different styles. It can also generate videos with different frame rates and resolutions during both training and inference phases, suitable for both images and videos. Moreover, we introduce slice VAE, a novel approach to condense the temporal axis, facilitating the generation of long duration videos. Currently, EasyAnimate exhibits the proficiency to generate videos with 144 frames. We provide a holistic ecosystem for video production based on DiT, encompassing aspects such as data pre-processing, VAE training, DiT models training (both the baseline model and LoRA model), and end-to-end video inference. Code is available at: https://github.com/aigc-apps/EasyAnimate. We are continuously working to enhance the performance of our method., Comment: 8 pages, 6 figures

Magnetars are potential energy sources or central engines for numerous transient phenomena in the Universe. How newborn magnetars evolve in different environments remains an open question. Based on both observed and candidate magnetars, it is found that the periods of all magnetars or candidates appear as a bimodal distribution, and are defined as the ``long-P'' and ``short-P'' magnetar subclasses, respectively. We find that for the ``short-P'' subclass of magnetars, the $\dot{P}$ values also appear as a bimodal distribution, and therefore can be classified as ``high-$\dot{P}$ short-P'' and ``low-$\dot{P}$ short-P'' magnetar subclasses. In this paper, we use Monte Carlo simulations to generate synthetic magnetar populations and investigate the evolution of the ``high-$\dot{P}$ short-P'' and ``low-$\dot{P}$ short-P'' magnetar subclasses by considering both the magnetar spin and inclination, as well as the decay of their magnetic field within their evolution in both vacuum and plasma-filled magnetospheres. We find that the magnetar evolution is dependent on both spin and magnetic field, but seems to be insensitive to inclination evolution and magnetospheric environment for the ``high-$\dot{P}$ short-P'' sub-class. In comparison for the case of ``high-$\dot{P}$ short-P'', the magnetar evolution is dependent on spin, magnetic field, and inclination evolution, as well as the magnetospheric environment. The best evolution model should be the case of inclination evolution in vacuum with a small value of $\overline{\mathrm{FOM}}$. The differences in the best-fit parameters also suggest that the ``high-$\dot{P}$ short-P'' and ``low-$\dot{P}$ short-P'' magnetar subclasses may be tracking with different evolution channels., Comment: 18 pages, 2 Tables, and 9 Figures. PRD in press, and matched with the published verison

Instruction tuning aims to align large language models (LLMs) with open-domain instructions and human-preferred responses. While several studies have explored autonomous approaches to distilling and annotating instructions from powerful proprietary LLMs, such as ChatGPT, they often neglect the impact of the distributions and characteristics of tasks, together with the varying difficulty of instructions in training sets. This oversight can lead to imbalanced knowledge capabilities and poor generalization powers of student LLMs. To address these challenges, we introduce Task-Aware Curriculum Planning for Instruction Refinement (TAPIR), a multi-round distillation framework that utilizes an oracle LLM to select instructions that are difficult for a student LLM to follow. To balance the student's capabilities, task distributions in training sets are adjusted with responses automatically refined according to their corresponding tasks. In addition, by incorporating curriculum planning, our approach systematically escalates the difficulty levels of tasks, progressively enhancing the student LLM's capabilities. We rigorously evaluate TAPIR using several widely recognized benchmarks (such as AlpacaEval 2.0, MT-Bench, etc.) and multiple student LLMs. Empirical results demonstrate that student LLMs, trained with our method and less training data, outperform larger instruction-tuned models and strong distillation baselines., Comment: emnlp 2024 findings

In visual tasks, large teacher models capture essential features and deep information, enhancing performance. However, distilling this information into smaller student models often leads to performance loss due to structural differences and capacity limitations. To tackle this, we propose a distillation framework based on graph knowledge, including a multi-level feature alignment strategy and an attention-guided mechanism to provide a targeted learning trajectory for the student model. We emphasize spectral embedding (SE) as a key technique in our distillation process, which merges the student's feature space with the relational knowledge and structural complexities similar to the teacher network. This method captures the teacher's understanding in a graph-based representation, enabling the student model to more accurately mimic the complex structural dependencies present in the teacher model. Compared to methods that focus only on specific distillation areas, our strategy not only considers key features within the teacher model but also endeavors to capture the relationships and interactions among feature sets, encoding these complex pieces of information into a graph structure to understand and utilize the dynamic relationships among these pieces of information from a global perspective. Experiments show that our method outperforms previous feature distillation methods on the CIFAR-100, MS-COCO, and Pascal VOC datasets, proving its efficiency and applicability.

Retrieval-augmented large language models (LLMs) leverage relevant content retrieved by information retrieval systems to generate correct responses, aiming to alleviate the hallucination problem. However, existing retriever-responder methods typically append relevant documents to the prompt of LLMs to perform text generation tasks without considering the interaction of fine-grained structural semantics between the retrieved documents and the LLMs. This issue is particularly important for accurate response generation as LLMs tend to "lose in the middle" when dealing with input prompts augmented with lengthy documents. In this work, we propose a new pipeline named "Reinforced Retriever-Reorder-Responder" (R$^4$) to learn document orderings for retrieval-augmented LLMs, thereby further enhancing their generation abilities while the large numbers of parameters of LLMs remain frozen. The reordering learning process is divided into two steps according to the quality of the generated responses: document order adjustment and document representation enhancement. Specifically, document order adjustment aims to organize retrieved document orderings into beginning, middle, and end positions based on graph attention learning, which maximizes the reinforced reward of response quality. Document representation enhancement further refines the representations of retrieved documents for responses of poor quality via document-level gradient adversarial learning. Extensive experiments demonstrate that our proposed pipeline achieves better factual question-answering performance on knowledge-intensive tasks compared to strong baselines across various public datasets. The source codes and trained models will be released upon paper acceptance., Comment: need to further experiment

This paper proposes a new gradient-based XAI method called Guided AbsoluteGrad for saliency map explanations. We utilize both positive and negative gradient magnitudes and employ gradient variance to distinguish the important areas for noise deduction. We also introduce a novel evaluation metric named ReCover And Predict (RCAP), which considers the Localization and Visual Noise Level objectives of the explanations. We propose two propositions for these two objectives and prove the necessity of evaluating them. We evaluate Guided AbsoluteGrad with seven gradient-based XAI methods using the RCAP metric and other SOTA metrics in three case studies: (1) ImageNet dataset with ResNet50 model; (2) International Skin Imaging Collaboration (ISIC) dataset with EfficientNet model; (3) the Places365 dataset with DenseNet161 model. Our method surpasses other gradient-based approaches, showcasing the quality of enhanced saliency map explanations through gradient magnitude., Comment: CAI2024 Camera-ready Submission