Search

Understanding the mechanisms behind neural network optimization is crucial for improving network design and performance. While various optimization techniques have been developed, a comprehensive understanding of the underlying principles that govern these techniques remains elusive. Specifically, the role of symmetry breaking, a fundamental concept in physics, has not been fully explored in neural network optimization. This gap in knowledge limits our ability to design networks that are both efficient and effective. Here, we propose the symmetry breaking hypothesis to elucidate the significance of symmetry breaking in enhancing neural network optimization. We demonstrate that a simple input expansion can significantly improve network performance across various tasks, and we show that this improvement can be attributed to the underlying symmetry breaking mechanism. We further develop a metric to quantify the degree of symmetry breaking in neural networks, providing a practical approach to evaluate and guide network design. Our findings confirm that symmetry breaking is a fundamental principle that underpins various optimization techniques, including dropout, batch normalization, and equivariance. By quantifying the degree of symmetry breaking, our work offers a practical technique for performance enhancement and a metric to guide network design without the need for complete datasets and extensive training processes., Comment: 29 pages, 8 figures

Graph Neural Networks (GNNs) with equivariant properties have achieved significant success in modeling complex dynamic systems and molecular properties. However, their expressiveness ability is limited by: (1) Existing methods often overlook the over-smoothing issue caused by traditional GNN models, as well as the gradient explosion or vanishing problems in deep GNNs. (2) Most models operate on first-order information, neglecting that the real world often consists of second-order systems, which further limits the model's representation capabilities. To address these issues, we propose the \textbf{Du}al \textbf{S}econd-order \textbf{E}quivariant \textbf{G}raph \textbf{O}rdinary Differential Equation (\method{}) for equivariant representation. Specifically, \method{} apply the dual second-order equivariant graph ordinary differential equations (Graph ODEs) on graph embeddings and node coordinates, simultaneously. Theoretically, we first prove that \method{} maintains the equivariant property. Furthermore, we provide theoretical insights showing that \method{} effectively alleviates the over-smoothing problem in both feature representation and coordinate update. Additionally, we demonstrate that the proposed \method{} mitigates the exploding and vanishing gradients problem, facilitating the training of deep multi-layer GNNs. Extensive experiments on benchmark datasets validate the superiority of the proposed \method{} compared to baselines.

Post-selecting output states in measurements can effectively amplify weak signals and improve precision. However, post-selection effects may also introduce unintended biases in precision measurements. Here, we investigate the influence of post-selection in the precision spectroscopy of the $2^3S - 2^3P$ transition of helium ($^4$He) using an atomic beam. We directly observe that post-selection based on atomic positions causes a shift in the measured transition frequency, amounting to approximately -55 kHz. After accounting for this post-selection shift, we obtain a corrected frequency of $276,764,094,712.45 \pm 0.86$ kHz for the $2^3S_1 - 2^3P_0$ transition. Combining this result with existing data for $^3$He, we derive a new value for the difference in squared nuclear charge radii, $\delta r^2 [r_{h}^{2} - r_{\alpha}^{2}] = 1.0733 \pm 0.0021$ fm$^2$. This value shows a $2.8\sigma$ deviation from measurements of muonic helium ions, potentially pointing to new physics that challenges lepton universality in quantum electrodynamics., Comment: 14 pages including appendix

Water's unique hydrogen-bonding network and anomalous properties present significant challenges for accurately modeling its structural, thermodynamic, and transport behavior across varied conditions. Although machine-learned potentials have advanced the prediction of individual properties, a unified computational framework capable of simultaneously capturing water's complex and subtle properties with high accuracy has remained elusive. Here, we address this challenge by introducing NEP-MB-pol, a highly accurate and efficient neuroevolution potential trained on extensive MB-pol reference data with coupled-cluster-level accuracy, combined with path-integral molecular dynamics and quantum-correction techniques to incorporate nuclear quantum effects. This NEP-MB-pol framework reproduces experimentally measured structural, thermodynamic, and transport properties of water across a broad temperature range, achieving simultaneous, fast, and accurate prediction of self-diffusion coefficient, viscosity, and thermal conductivity. Our approach provides a unified and robust tool for exploring thermodynamic and transport properties of water under diverse conditions, with significant potential for broader applications across research fields., Comment: 12 pages, 4 figures in the main text; 8 figures in the SI

Stuttering is a neurodevelopmental speech disorder characterized by common speech symptoms such as pauses, exclamations, repetition, and prolongation. Speech-language pathologists typically assess the type and severity of stuttering by observing these symptoms. Many effective end-to-end methods exist for stuttering detection, but a commonly overlooked challenge is the uncertain relationship between tasks involved in this process. Using a suitable multi-task strategy could improve stuttering detection performance. This paper presents a novel stuttering event detection model designed to help speech-language pathologists assess both the type and severity of stuttering. First, the Conformer model extracts acoustic features from stuttered speech, followed by a Long Short-Term Memory (LSTM) network to capture contextual information. Finally, we explore multi-task learning for stuttering and propose an effective multi-task strategy. Experimental results show that our model outperforms current state-of-the-art methods for stuttering detection. In the SLT 2024 Stuttering Speech Challenge based on the AS-70 dataset [1], our model improved the mean F1 score by 24.8% compared to the baseline method and achieved first place. On this basis, we conducted relevant extensive experiments on LSTM and multi-task learning strategies respectively. The results show that our proposed method improved the mean F1 score by 39.8% compared to the baseline method., Comment: 7 pages, 3 figures, 7 tables

DarkSHINE is a newly proposed fixed-target experiment initiative to search for the invisible decay of Dark Photon via missing energy/momentum signatures, based on the high repetition rate electron beam to be deployed/delivered by the Shanghai High repetition rate XFEL and Extreme light facility (SHINE). This report elaborates the baseline design of DarkSHINE experiment by introducing the physics goals, experimental setups, details of each sub-detector system technical designs, signal and backgground modelings, expected search sensitivities and future prospects, which mark an important step towards the further prototyping and technical demonstrations.

In this paper we investigate the integrable boundary state in ABJM theory. We find an integrability condition for the two-site integrable matrix product state (MPS) similar to the KT-relation. We also construct a class of non-trivial MPSs from the projected K-matrices.

We assume the coupling of Axion-like particle (ALP) to diphoton $g_{a\gamma\gamma} \sim 0$ and accomplish detailed analyses for the $\gamma Z$ decay mode of such heavy photophobic ALPs at the high luminosity-Large Hadron Collider (HL-LHC). ALPs are produced with two jets via both the $s$-channel vector boson exchange and vector boson fusion processes, with the signal process $pp \to jj\, a (\to \gamma\, Z (\to \ell^+ \ell^-)\, ) $ for $\ell = e, \mu$. Signal and background events are simulated at the detector-level. Preselection criteria target events with one photon, two oppositely charged electrons or muons, and two non-$b$-tagged jets. The ALP mass is reconstructed, and kinematic observables are input into a machine learning-based multivariate analysis for optimal background rejection. Discover sensitivities on the ALP's coupling to di-$W$ boson, $g_{aWW}$, are presented in the mass range from 100 to 4000 GeV at center-of-mass energy $\sqrt{s} = 14$ TeV and integrated luminosity $\mathcal{L} =$ 3 ab$^{-1}$ and 140 fb$^{-1}$. Sensitivities on the production cross section $\sigma (pp \to jj\, a)$ times the branching ratio Br$(a \to \gamma Z)$ are also presented., Comment: 22 pages, 8 figures, 3 tables

Quantum mechanics gives a new breakthrough to the field of parameter estimation. In the realm of quantum metrology, the precision of parameter estimation is limited by the quantum Fisher information. We introduce the measures of partial coherence based on (quantum) Fisher information by taking into account the post-selective non-unitary parametrization process. These partial coherence measures present a clear operational interpretation by directly linking the coherence to the parameter estimation accuracy. Furthermore, we explore the distinctions between our partial coherence measure and the quantum Fisher information within the context of unitary parametrization. We provide an analytical expression for the partial coherence measure of two-qubit states. We elucidate the operational significance of the partial coherence measures by establishing the connections between the partial coherence measures and quantum state discrimination., Comment: 17 pages, 1 figure

The main goal of this paper is to investigate the multi-parameter stability result for a stochastic fractional differential variational inequality with L\'{e}vy jump (SFDVI with L\'{e}vy jump) under some mild conditions. We verify that Mosco convergence of the perturbed set implies point convergence of the projection onto the Hilbert space consisting of special stochastic processes whose range is the perturbed set. Moreover, by using the projection method and some inequality techniques, we establish a strong convergence result for the solution of SFDVI with L\'{e}vy jump when the mappings and constraint set are both perturbed. Finally, we apply the stability results to the spatial price equilibrium problem and the multi-agent optimization problem in stochastic environments.

Soft- and hard-constrained Physics Informed Neural Networks (PINNs) have achieved great success in solving partial differential equations (PDEs). However, these methods still face great challenges when solving the Navier-Stokes equations (NSEs) with complex boundary conditions. To address these challenges, this paper introduces a novel complementary scheme combining soft and hard constraint PINN methods. The soft-constrained part is thus formulated to obtain the preliminary results with a lighter training burden, upon which refined results are then achieved using a more sophisticated hard-constrained mechanism with a primary network and a distance metric network. Specifically, the soft-constrained part focuses on boundary points, while the primary network emphasizes inner domain points, primarily through PDE loss. Additionally, the novel distance metric network is proposed to predict the power function of the distance from a point to the boundaries, which serves as the weighting factor for the first two components. This approach ensures accurate predictions for both boundary and inner domain areas. The effectiveness of the proposed method on the NSEs problem with complex boundary conditions is demonstrated by solving a 2D cylinder wake problem and a 2D blocked cavity flow with a segmented inlet problem, achieving significantly higher accuracy compared to traditional soft- and hard-constrained PINN approaches. Given PINN's inherent advantages in solving the inverse and the large-scale problems, which are challenging for traditional computational fluid dynamics (CFD) methods, this approach holds promise for the inverse design of required flow fields by specifically-designed boundary conditions and the reconstruction of large-scale flow fields by adding a limited number of training input points. The code for our approach will be made publicly available.

While large language models (LLMs) have demonstrated increasing power, they have also called upon studies on their hallucinated outputs that deviate from factually correct statements. In this paper, we focus on one important scenario of false premises, where LLMs are distracted by misaligned claims although the model possesses the required factual knowledge to answer original questions accurately. Inspired by the observation that entropy of the false-premise prompt is closely related to its likelihood to elicit hallucination generation, we propose a new prompting algorithm, named DecoPrompt, to mitigate hallucination. DecoPrompt leverages LLMs to "decode" the false-premise prompts without really eliciting hallucination output from LLMs. We perform experiments on two datasets, demonstrating that DecoPrompt can reduce hallucinations effectively on outputs from different LLMs. Moreover, DecoPrompt exhibits cross-model transferability, which facilitates its applications to scenarios such as LLMs of large sizes or unavailable model logits.

Large Language Models (LLMs) have achieved significant success in various natural language processing tasks, but the role of wireless networks in supporting LLMs has not been thoroughly explored. In this paper, we propose a wireless distributed Mixture of Experts (WDMoE) architecture to enable collaborative deployment of LLMs across edge servers at the base station (BS) and mobile devices in wireless networks. Specifically, we decompose the MoE layer in LLMs by placing the gating network and the preceding neural network layer at BS, while distributing the expert networks among the devices. This deployment leverages the parallel inference capabilities of expert networks on mobile devices, effectively utilizing the limited computing and caching resources of these devices. Accordingly, we develop a performance metric for WDMoE-based LLMs, which accounts for both model capability and latency. To minimize the latency while maintaining accuracy, we jointly optimize expert selection and bandwidth allocation based on the performance metric. Moreover, we build a hardware testbed using NVIDIA Jetson kits to validate the effectiveness of WDMoE. Both theoretical simulations and practical hardware experiments demonstrate that the proposed method can significantly reduce the latency without compromising LLM performance.

Node classification with Graph Neural Networks (GNN) under a fixed set of labels is well known in contrast to Graph Few-Shot Class Incremental Learning (GFSCIL), which involves learning a GNN classifier as graph nodes and classes growing over time sporadically. We introduce inductive GFSCIL that continually learns novel classes with newly emerging nodes while maintaining performance on old classes without accessing previous data. This addresses the practical concern of transductive GFSCIL, which requires storing the entire graph with historical data. Compared to the transductive GFSCIL, the inductive setting exacerbates catastrophic forgetting due to inaccessible previous data during incremental training, in addition to overfitting issue caused by label sparsity. Thus, we propose a novel method, called Topology-based class Augmentation and Prototype calibration (TAP). To be specific, it first creates a triple-branch multi-topology class augmentation method to enhance model generalization ability. As each incremental session receives a disjoint subgraph with nodes of novel classes, the multi-topology class augmentation method helps replicate such a setting in the base session to boost backbone versatility. In incremental learning, given the limited number of novel class samples, we propose an iterative prototype calibration to improve the separation of class prototypes. Furthermore, as backbone fine-tuning poses the feature distribution drift, prototypes of old classes start failing over time, we propose the prototype shift method for old classes to compensate for the drift. We showcase the proposed method on four datasets.

Studying the periodic flux-variation behavior of blazars is vital for probing supermassive black hole binaries and the kinematics of relativistic jets. In this work, we report the detection of the multi-band possible periodic variations of the blazar PKS J2134-0153, including the infrared ($1.6(\pm0.4)\times 10^3$ days) and optical ($1.8(\pm1)\times 10^3$ days). The periods in the infrared and optical bands are statistically consistent with the period in the radio band ($P_{\mathrm{Radio}}$$ = 1760\pm33$ days, obtained from our previous work). Moreover, flux variations in different bands are correlated with evident inter-band time delays, and the time lags of infrared and optical emission with respect to radio emission are $(3.3\pm2.3)\times10^{2}$ days and $(3.0\pm2.3)\times10^{2}$ days, respectively. The cross-correlations indicate a common origin of radio, infrared, and optical emission. The relative positions between emission regions of infrared and optical emission to radio emission are estimated according to the time lags, i.e., $0.37\pm0.26$ pc and $0.33\pm0.26$ pc. The relative distances seem to be quantitatively consistent with the theoretical prediction., Comment: Accepted to MNRAS

The core of self-supervised point cloud learning lies in setting up appropriate pretext tasks, to construct a pre-training framework that enables the encoder to perceive 3D objects effectively. In this paper, we integrate two prevalent methods, masked point modeling (MPM) and 3D-to-2D generation, as pretext tasks within a pre-training framework. We leverage the spatial awareness and precise supervision offered by these two methods to address their respective limitations: ambiguous supervision signals and insensitivity to geometric information. Specifically, the proposed framework, abbreviated as PointCG, consists of a Hidden Point Completion (HPC) module and an Arbitrary-view Image Generation (AIG) module. We first capture visible points from arbitrary views as inputs by removing hidden points. Then, HPC extracts representations of the inputs with an encoder and completes the entire shape with a decoder, while AIG is used to generate rendered images based on the visible points' representations. Extensive experiments demonstrate the superiority of the proposed method over the baselines in various downstream tasks. Our code will be made available upon acceptance.

This study establishes a skier-ski-snow interaction (SSSI) model that integrates a 3D full-body musculoskeletal model, a flexible ski model, a ski-snow contact model, and an air resistance model. An experimental method is developed to collect kinematic and kinetic data using IMUs, GPS, and plantar pressure measurement insoles, which are cost-effective and capable of capturing motion in large-scale field conditions. The ski-snow interaction parameters are optimized for dynamic alignment with snow conditions and individual turning techniques. Forward-inverse dynamics simulation is performed using only the skier's posture as model input and leaving the translational degrees of freedom (DOFs) between the pelvis and the ground unconstrained. The effectiveness of our model is further verified by comparing the simulated results with the collected GPS and plantar pressure data. The correlation coefficient between the simulated ski-snow contact force and the measured plantar pressure data is 0.964, and the error between the predicted motion trajectory and GPS data is 0.7%. By extracting kinematic and kinetic parameters from skiers of different skill levels, quantitative performance analysis helps quantify ski training. The SSSI model with the parameter optimization algorithm of the ski-snow interaction allows for the description of skiing characteristics across varied snow conditions and different turning techniques, such as carving and skidding. Our research advances the understanding of alpine skiing dynamics, informing the development of training programs and facility designs to enhance athlete performance and safety.

With the versatile manipulation capability, programmable metasurfaces are rapidly advancing in their intelligence, integration, and commercialization levels. However, as the programmable metasurfaces scale up, their control configuration becomes increasingly complicated, posing significant challenges and limitations. Here, we propose a multiple-partition cross-modulation (MPCM) programmable metasurface to enhance the wireless communication coverage with low hardware complexity. We firstly propose an innovative encoding scheme to multiply the control voltage vectors of row-column crossing, achieving high beamforming precision in free space while maintaining low control hardware complexity and reducing memory requirements for coding sequences. We then design and fabricate an MPCM programmable metasurface to confirm the effectiveness of the proposed encoding scheme. The simulated and experimental results show good agreements with the theoretically calculated outcomes in beam scanning across the E and H planes and in free-space beam pointing. The MPCM programmable metasurface offers strong flexibility and low complexity by allowing various numbers and combinations of partition items in modulation methods, catering to diverse precision demands in various scenarios. We demonstrate the performance of MPCM programmable metasurface in a realistic indoor setting, where the transmissions of videos to specific receiver positions are successfully achieved, surpassing the capabilities of traditional programmable metasurfaces. We believe that the proposed programmable metasurface has great potentials in significantly empowering the wireless communications while addressing the challenges associated with the programmable metasurface's design and implementation.

Lifelong few-shot customization for text-to-image diffusion aims to continually generalize existing models for new tasks with minimal data while preserving old knowledge. Current customization diffusion models excel in few-shot tasks but struggle with catastrophic forgetting problems in lifelong generations. In this study, we identify and categorize the catastrophic forgetting problems into two folds: relevant concepts forgetting and previous concepts forgetting. To address these challenges, we first devise a data-free knowledge distillation strategy to tackle relevant concepts forgetting. Unlike existing methods that rely on additional real data or offline replay of original concept data, our approach enables on-the-fly knowledge distillation to retain the previous concepts while learning new ones, without accessing any previous data. Second, we develop an In-Context Generation (ICGen) paradigm that allows the diffusion model to be conditioned upon the input vision context, which facilitates the few-shot generation and mitigates the issue of previous concepts forgetting. Extensive experiments show that the proposed Lifelong Few-Shot Diffusion (LFS-Diffusion) method can produce high-quality and accurate images while maintaining previously learned knowledge.

Quantum entanglement is a unique correlation phenomenon in quantum mechanics, and the measurement of quantum entanglement plays an important role in quantum computing and quantum communication. Many mainstream entanglement criteria and measurement methods currently known have shortcomings in certain aspects, such as not being sufficient or necessary conditions for entanglement, or only being effective in simple cases such as 2-qubits or pure states. In this work, we will propose a scheme for measuring quantum entanglement, which starts with treating the set of quantum separable states as a convex hull of quantum separable pure states, and analyzes the properties of the convex hull to obtain a new form of entanglement measurement. Although a large amount of data is required in the measurement process, this method is not only applicable to 2-qubit quantum states, but also a entanglement measurement method that can be applied to any dimension and any fragment. We will provide several examples to compare their results with other entanglement metrics and entanglement determination methods to verify their feasibility.

Data assimilation (DA) combines partial observations with a dynamical model to improve state estimation. Filter-based DA uses only past and present data and is the prerequisite for real-time forecasts. Smoother-based DA exploits both past and future observations. It aims to fill in missing data, provide more accurate estimations, and develop high-quality datasets. However, the standard smoothing procedure requires using all historical state estimations, which is storage-demanding, especially for high-dimensional systems. This paper develops an adaptive-lag online smoother for a large class of complex dynamical systems with strong nonlinear and non-Gaussian features, which has important applications to many real-world problems. The adaptive lag allows the DA to utilize only observations within a nearby window, significantly reducing computational storage. Online lag adjustment is essential for tackling turbulent systems, where temporal autocorrelation varies significantly over time due to intermittency, extreme events, and nonlinearity. Based on the uncertainty reduction in the estimated state, an information criterion is developed to systematically determine the adaptive lag. Notably, the mathematical structure of these systems facilitates the use of closed analytic formulae to calculate the online smoother and the adaptive lag, avoiding empirical tunings as in ensemble-based DA methods. The adaptive online smoother is applied to studying three important scientific problems. First, it helps detect online causal relationships between state variables. Second, its advantage of computational storage is illustrated via Lagrangian DA, a high-dimensional nonlinear problem. Finally, the adaptive smoother advances online parameter estimation with partial observations, emphasizing the role of the observed extreme events in accelerating convergence., Comment: 40 pages, 7 figures, typeset in LaTeX. Submitted for peer-review to Springer Nature's Journal of Nonlinear Science. For more info see https://sites.google.com/wisc.edu/mariosandreou/pubs-and-talks/cgns-online-martingale-free#h.55a05qfs9w12

Understanding the interactions between the El Nino-Southern Oscillation (ENSO) and the Madden-Julian Oscillation (MJO) is essential to studying climate variabilities and predicting extreme weather events. Here, we develop a stochastic conceptual model for describing the coupled ENSO-MJO phenomenon. The model adopts a three-box representation of the interannual ocean component to characterize the ENSO diversity. For the intraseasonal atmospheric component, a low-order Fourier representation is used to describe the eastward propagation of the MJO. We incorporate decadal variability to account for modulations in the background state that influence the predominant types of El Nino events. In addition to dynamical coupling through wind forcing and latent heat, state-dependent noise is introduced to characterize the statistical interactions among these multiscale processes, improving the simulation of extreme events. The model successfully reproduces the observed non-Gaussian statistics of ENSO diversity and MJO spectra. It also captures the interactions between wind, MJO, and ENSO., Comment: 26 pages, 12 figures

This paper introduces a vibrotactile belt for interpersonal synchronization of breath. It can synchronize the breathing tempo of two people by transferring breathing rhythm of one user to vibration signals of another belt, where the depth of breathing is represented by the intensity of vibration. This provides a novel way of emotional connect between people. Meanwhile, this breath-sharing device may also be combined with smart devices in the future to form a one-to-many, many-to-many internet of breath, which has promising application prospects in healthcare, sports breathing guidance and other scenarios., Comment: Part of proceedings of 6th International Conference AsiaHaptics 2024

Recent advances in large language models (LLMs), make it potentially feasible to automatically refactor source code with LLMs. However, it remains unclear how well LLMs perform compared to human experts in conducting refactorings automatically and accurately. To fill this gap, in this paper, we conduct an empirical study to investigate the potential of LLMs in automated software refactoring, focusing on the identification of refactoring opportunities and the recommendation of refactoring solutions. We first construct a high-quality refactoring dataset comprising 180 real-world refactorings from 20 projects, and conduct the empirical study on the dataset. With the to-be-refactored Java documents as input, ChatGPT and Gemini identified only 28 and 7 respectively out of the 180 refactoring opportunities. However, explaining the expected refactoring subcategories and narrowing the search space in the prompts substantially increased the success rate of ChatGPT from 15.6% to 86.7%. Concerning the recommendation of refactoring solutions, ChatGPT recommended 176 refactoring solutions for the 180 refactorings, and 63.6% of the recommended solutions were comparable to (even better than) those constructed by human experts. However, 13 out of the 176 solutions suggested by ChatGPT and 9 out of the 137 solutions suggested by Gemini were unsafe in that they either changed the functionality of the source code or introduced syntax errors, which indicate the risk of LLM-based refactoring. To this end, we propose a detect-and-reapply tactic, called RefactoringMirror, to avoid such unsafe refactorings. By reapplying the identified refactorings to the original code using thoroughly tested refactoring engines, we can effectively mitigate the risks associated with LLM-based automated refactoring while still leveraging LLM's intelligence to obtain valuable refactoring recommendations.

In this era of videos, automatic video editing techniques attract more and more attention from industry and academia since they can reduce workloads and lower the requirements for human editors. Existing automatic editing systems are mainly scene- or event-specific, e.g., soccer game broadcasting, yet the automatic systems for general editing, e.g., movie or vlog editing which covers various scenes and events, were rarely studied before, and converting the event-driven editing method to a general scene is nontrivial. In this paper, we propose a two-stage scheme for general editing. Firstly, unlike previous works that extract scene-specific features, we leverage the pre-trained Vision-Language Model (VLM) to extract the editing-relevant representations as editing context. Moreover, to close the gap between the professional-looking videos and the automatic productions generated with simple guidelines, we propose a Reinforcement Learning (RL)-based editing framework to formulate the editing problem and train the virtual editor to make better sequential editing decisions. Finally, we evaluate the proposed method on a more general editing task with a real movie dataset. Experimental results demonstrate the effectiveness and benefits of the proposed context representation and the learning ability of our RL-based editing framework.

Audio-Visual Question Answering (AVQA) is a challenging task that involves answering questions based on both auditory and visual information in videos. A significant challenge is interpreting complex multi-modal scenes, which include both visual objects and sound sources, and connecting them to the given question. In this paper, we introduce the Source-aware Semantic Representation Network (SaSR-Net), a novel model designed for AVQA. SaSR-Net utilizes source-wise learnable tokens to efficiently capture and align audio-visual elements with the corresponding question. It streamlines the fusion of audio and visual information using spatial and temporal attention mechanisms to identify answers in multi-modal scenes. Extensive experiments on the Music-AVQA and AVQA-Yang datasets show that SaSR-Net outperforms state-of-the-art AVQA methods., Comment: EMNLP 2024

The eXTP mission is a major project of the Chinese Academy of Sciences (CAS), with a large involvement of Europe. Its scientific payload includes four instruments: SFA, PFA, LAD and WFM. They offer an unprecedented simultaneous wide-band Xray timing and polarimetry sensitivity. A large European consortium is contributing to the eXTP study, both for the science and the instrumentation. Europe is expected to provide two of the four instruments: LAD and WFM; the LAD is led by Italy and the WFM by Spain. The WFM for eXTP is based on the design originally proposed for the LOFT ESA M3 mission, that underwent a Phase A feasibility study. It will be a wide field of view X-ray monitor instrument working in the 2-50 keV energy range, achieved with large-area Silicon Drift Detectors (SDDs), similar to the ones used for the LAD but with better spatial resolution. The WFM will consist of 3 pairs of coded mask cameras with a total combined field of view (FoV) of 90x180 degrees at zero response and a source localisation accuracy of ~1 arc min. The main goal of the WFM is to provide triggers for the target of opportunity observations of the SFA, PFA and LAD, in order to perform the core science programme, dedicated to the study of matter under extreme conditions of density, gravity and magnetism. In addition, the unprecedented combination of large field of view and imaging capability, down to 2 keV, of the WFM will allow eXTP to make important discoveries of the variable and transient X-ray sky, and provide X-ray coverage of a broad range of astrophysical objects covered under 'observatory science', such as gamma-ray bursts, fast radio bursts, gravitational wave electromagnetic counterparts. In this paper we provide an overview of the WFM instrument, explaining its design, configuration, and anticipated performance., Comment: 15 pages, 13 figures, Proceedings of SPIE 13093, Space Telescopes and Instrumentation 2024: Ultraviolet to Gamma Ray; Proceedings Volume 13093, Space Telescopes and Instrumentation 2024: Ultraviolet to Gamma Ray; 130931Y (2024); doi: 10.1117/12.3020020

In this paper, we introduce a novel continual audio-visual sound separation task, aiming to continuously separate sound sources for new classes while preserving performance on previously learned classes, with the aid of visual guidance. This problem is crucial for practical visually guided auditory perception as it can significantly enhance the adaptability and robustness of audio-visual sound separation models, making them more applicable for real-world scenarios where encountering new sound sources is commonplace. The task is inherently challenging as our models must not only effectively utilize information from both modalities in current tasks but also preserve their cross-modal association in old tasks to mitigate catastrophic forgetting during audio-visual continual learning. To address these challenges, we propose a novel approach named ContAV-Sep (\textbf{Cont}inual \textbf{A}udio-\textbf{V}isual Sound \textbf{Sep}aration). ContAV-Sep presents a novel Cross-modal Similarity Distillation Constraint (CrossSDC) to uphold the cross-modal semantic similarity through incremental tasks and retain previously acquired knowledge of semantic similarity in old models, mitigating the risk of catastrophic forgetting. The CrossSDC can seamlessly integrate into the training process of different audio-visual sound separation frameworks. Experiments demonstrate that ContAV-Sep can effectively mitigate catastrophic forgetting and achieve significantly better performance compared to other continual learning baselines for audio-visual sound separation. Code is available at: \url{https://github.com/weiguoPian/ContAV-Sep_NeurIPS2024}., Comment: NeurIPS 2024

As demands for communication and computational capabilities escalate, traditional bit-oriented communication falls short of these stringent requirements, especially for mission-critical and computation-intensive applications. Visual Question Answering (VQA), a representative application, has adopted edge computing to mitigate local computational constraints and accelerate visual perception with natural language. However, it encounters significant communication challenges such as limited bandwidth, reduced transmission power, and increased noise levels, leading to considerable latency and reduced efficiency in image and question transmission. we propose a goal-oriented semantic communication (GSC) framework that focuses on effectively extracting and transmitting semantic information most relevant to the VQA goals, improving the answering accuracy and enhancing the effectiveness and efficiency. The objective is to maximize the answering accuracy, and we propose a scene graphs (SG)-based image semantic extraction and ranking approach to prioritize the semantic information based on the goal of questions. Experimental results demonstrate that our GSC framework improves answering accuracy by up to 59% under Rayleigh channels while reducing total latency by up to 65% compared to traditional bit-oriented transmission.

The streaming bipartite graph is extensively used to model the dynamic relationship between two types of entities in many real-world applications, such as movie recommendations, location-based services, and online shopping. Since it contains abundant information, discovering the dense subgraph with high structural cohesiveness (i.e., community detection) in the bipartite streaming graph is becoming a valuable problem. Inspired by this, in this paper, we study the structure of community on the butterfly motif in the bipartite graph. We propose a novel problem, named Community Detection over Streaming Bipartite Network (CD-SBN), which aims to retrieve qualified communities with user-specific query keywords and high structural cohesiveness at snapshot and continuous scenarios. In particular, we formulate the user relationship score in the weighted bipartite network via the butterfly pattern and define a novel $(k,r,\sigma)$-bitruss as the community structure. To efficiently tackle the CD-SBN problem, we design effective pruning strategies to rule out false alarms of $(k,r,\sigma)$-bitruss and propose a hierarchical synopsis to facilitate the CD-SBN processing. Due to the dynamic of streaming bipartite networks, we devise an efficient procedure for incremental graph maintenance. We develop an efficient algorithm to answer the snapshot and continuous CD-SBN query by traversing the synopsis and applying the pruning strategies. With extensive experiments, we demonstrate the efficiency and effectiveness of our proposed CD-SBN processing approach over real/synthetic streaming bipartite networks.

The Water Cherenkov Detector Array (WCDA) is one of the components of Large High Altitude Air Shower Observatory (LHAASO) and can monitor any sources over two-thirds of the sky for up to 7 hours per day with >98\% duty cycle. In this work, we report the detection of two outbursts of the Fanaroff-Riley I radio galaxy NGC 1275 that were detected by LHAASO-WCDA between November 2022 and January 2023 with statistical significance of 5.2~$\sigma$ and 8.3~$\sigma$. The observed spectral energy distribution in the range from 500 GeV to 3 TeV is fitted by a power-law with a best-fit spectral index of $\alpha=-3.37\pm0.52$ and $-3.35\pm0.29$, respectively. The outburst flux above 0.5~TeV was ($4.55\pm 4.21)\times~10^{-11}~\rm cm^{-2}~s^{-1}$ and ($3.45\pm 1.78)\times~10^{-11}~\rm cm^{-2}~s^{-1}$, corresponding to 60\%, 45\% of Crab Nebula flux. Variation analysis reveals the variability time-scale of days at the TeV energy band. A simple test by one-zone synchrotron self-Compton model reproduces the data in the gamma-ray band well., Comment: 11 pages, 8 figures, 3 tables

Retrieval-Augmented Generation (RAG) has emerged as a reliable external knowledge augmentation technique to mitigate hallucination issues and parameterized knowledge limitations in Large Language Models (LLMs). Existing Adaptive RAG (ARAG) systems struggle to effectively explore multiple retrieval sources due to their inability to select the right source at the right time. To address this, we propose a multi-source ARAG framework, termed MSPR, which synergizes reasoning and preference-driven retrieval to adaptive decide "when and what to retrieve" and "which retrieval source to use". To better adapt to retrieval sources of differing characteristics, we also employ retrieval action adjustment and answer feedback strategy. They enable our framework to fully explore the high-quality primary source while supplementing it with secondary sources at the right time. Extensive and multi-dimensional experiments conducted on three datasets demonstrate the superiority and effectiveness of MSPR., Comment: 5 pages, 1 figure

We present Fashion-VDM, a video diffusion model (VDM) for generating virtual try-on videos. Given an input garment image and person video, our method aims to generate a high-quality try-on video of the person wearing the given garment, while preserving the person's identity and motion. Image-based virtual try-on has shown impressive results; however, existing video virtual try-on (VVT) methods are still lacking garment details and temporal consistency. To address these issues, we propose a diffusion-based architecture for video virtual try-on, split classifier-free guidance for increased control over the conditioning inputs, and a progressive temporal training strategy for single-pass 64-frame, 512px video generation. We also demonstrate the effectiveness of joint image-video training for video try-on, especially when video data is limited. Our qualitative and quantitative experiments show that our approach sets the new state-of-the-art for video virtual try-on. For additional results, visit our project page: https://johannakarras.github.io/Fashion-VDM., Comment: Accepted to SIGGRAPH Asia 2024

We perform a comprehensive bottom-up study of all the simplest lepton models based on non-holomorphic $A_{5}$ modular flavor symmetry, in which neutrinos are assumed to be Majorana particles and their masses are generated by the Weinberg operator or the type I seesaw mechanism. In the case that the generalized CP (gCP) symmetry is not considered, we find that 21 Weinberg operator models and 174 seesaw models can accommodate the experimental data in lepton sector, and all of them depend on six dimensionless free parameters and two overall scales. If gCP symmetry compatible with $A_{5}$ modular symmetry is imposed, one more free parameter would be reduced. Then only 4 of the 21 Weinberg operator models and 100 of the 174 seesaw models agree with the experimental data on lepton masses and mixing parameters. Furthermore, we perform a detailed numerical analysis for two example models for illustration., Comment: 45 pages, 5 figures

The Conditional Gaussian Nonlinear System (CGNS) is a broad class of nonlinear stochastic dynamical systems. Given the trajectories for a subset of state variables, the remaining follow a Gaussian distribution. Despite the conditionally linear structure, the CGNS exhibits strong nonlinearity, thus capturing many non-Gaussian characteristics observed in nature through its joint and marginal distributions. Desirably, it enjoys closed analytic formulae for the time evolution of its conditional Gaussian statistics, which facilitate the study of data assimilation and other related topics. In this paper, we develop a martingale-free approach to improve the understanding of CGNSs. This methodology provides a tractable approach to proving the time evolution of the conditional statistics by deriving results through time discretization schemes, with the continuous-time regime obtained via a formal limiting process as the discretization time-step vanishes. This discretized approach further allows for developing analytic formulae for optimal posterior sampling of unobserved state variables with correlated noise. These tools are particularly valuable for studying extreme events and intermittency and apply to high-dimensional systems. Moreover, the approach improves the understanding of different sampling methods in characterizing uncertainty. The effectiveness of the framework is demonstrated through a physics-constrained, triad-interaction climate model with cubic nonlinearity and state-dependent cross-interacting noise., Comment: 61 pages, 3 figures, typeset in LaTeX. Submitted for peer-review to MPDI's Entropy journal. For more info see https://sites.google.com/wisc.edu/mariosandreou/pubs-and-talks/cgns-online-martingale-free#h.jq7kwsq05ae2

In this paper, we propose a flexible SLAM framework, XRDSLAM. It adopts a modular code design and a multi-process running mechanism, providing highly reusable foundational modules such as unified dataset management, 3d visualization, algorithm configuration, and metrics evaluation. It can help developers quickly build a complete SLAM system, flexibly combine different algorithm modules, and conduct standardized benchmarking for accuracy and efficiency comparison. Within this framework, we integrate several state-of-the-art SLAM algorithms with different types, including NeRF and 3DGS based SLAM, and even odometry or reconstruction algorithms, which demonstrates the flexibility and extensibility. We also conduct a comprehensive comparison and evaluation of these integrated algorithms, analyzing the characteristics of each. Finally, we contribute all the code, configuration and data to the open-source community, which aims to promote the widespread research and development of SLAM technology within the open-source ecosystem.

Sea ice plays a crucial role in the climate system, particularly in the Marginal Ice Zone (MIZ), a transitional area consisting of fragmented ice between the open ocean and consolidated pack ice. As the MIZ expands, understanding its dynamics becomes essential for predicting climate change impacts. However, the role of clouds in these processes has been largely overlooked. This paper addresses that gap by developing an idealized coupled atmosphere-ocean-ice model incorporating cloud and precipitation effects, tackling both forward (simulation) and inverse (data assimilation) problems. Sea ice dynamics are modeled using the discrete element method, which simulates floes driven by atmospheric and oceanic forces. The ocean is represented by a two-layer quasi-geostrophic (QG) model, capturing mesoscale eddies and ice-ocean drag. The atmosphere is modeled using a two-layer saturated precipitating QG system, accounting for variable evaporation over sea surfaces and ice. Cloud cover affects radiation, influencing ice melting. The idealized coupled modeling framework allows us to study the interactions between atmosphere, ocean, and sea ice floes. Specifically, it focuses on how clouds and precipitation affect energy balance, melting, and freezing processes. It also serves as a testbed for data assimilation, which allows the recovery of unobserved floe trajectories and ocean fields in cloud-induced uncertainties. Numerical results show that appropriate reduced-order models help improve data assimilation efficiency with partial observations, allowing the skillful inference of missing floe trajectories and lower atmospheric winds. These results imply the potential of integrating idealized models with data assimilation to improve our understanding of Arctic dynamics and predictions.

We study the 7D Symmetry Topological Field Theory (SymTFT) associated to a 6D SCFT from the IIB/F-theory geometric engineering approach. The 6D (2,0) or (1,0) SCFT is constructed from IIB on a non-compact complex surface possibly with 7-branes. We derive the general form of 7D SymTFT actions from the compactification of IIB action on the boundary link of the base manifold of an elliptic Calabi-Yau threefold, for both the cases with or without flavor 7-branes intersecting the boundary link. Along the way we found new terms in the SymTFT action from the worldvolume action of flavor 7-branes involving the flavor center symmetries. We crosscheck the results against those obtained from either holographic constructions or the dual M-theory picture. Our construction potentially leads to a classification of the 7D SymTFTs which parallels the known geometric classification of the 6D SCFTs.

This paper investigates distributed computing and cooperative control of connected and automated vehicles (CAVs) in ramp merging scenario under transportation cyber-physical system. Firstly, a centralized cooperative trajectory planning problem is formulated subject to the safely constraints and traffic performance in ramp merging scenario, where the trajectories of all vehicles are jointly optimized. To get rid of the reliance on a central controller and reduce computation time, a distributed solution to this problem implemented among CAVs through Vehicles-to-Everything (V2X) communication is proposed. Unlike existing method, our method can distribute the computational task among CAVs and carry out parallel solving through V2X communication. Then, a multi-vehicles model predictive control (MPC) problem aimed at maximizing system stability and minimizing control input is formulated based on the solution of the first problem subject to strict safety constants and input limits. Due to these complex constraints, this problem becomes high-dimensional, centralized, and non-convex. To solve it in a short time, a decomposition and convex reformulation method, namely distributed cooperative iterative model predictive control (DCIMPC), is proposed. This method leverages the communication capability of CAVs to decompose the problem, making full use of the computational resources on vehicles to achieve fast solutions and distributed control. The two above problems with their corresponding solving methods form the systemic framework of the V2X assisted distributed computing and control. Simulations have been conducted to evaluate the framework's convergence, safety, and solving speed. Additionally, extra experiments are conducted to validate the performance of DCIMPC. The results show that our method can greatly improve computation speed without sacrificing system performance., Comment: This paper has been submitted to IEEE Journal. The source code has been released at: https://github.com/qiongwu86/V2X-Assisted-Distributed-Computing-and-Control-Framework-for-Connected-and-Automated-Vehicles.git

Event cameras provide a natural and data efficient representation of visual information, motivating novel computational strategies towards extracting visual information. Inspired by the biological vision system, we propose a behavior driven approach for object-wise distance estimation from event camera data. This behavior-driven method mimics how biological systems, like the human eye, stabilize their view based on object distance: distant objects require minimal compensatory rotation to stay in focus, while nearby objects demand greater adjustments to maintain alignment. This adaptive strategy leverages natural stabilization behaviors to estimate relative distances effectively. Unlike traditional vision algorithms that estimate depth across the entire image, our approach targets local depth estimation within a specific region of interest. By aligning events within a small region, we estimate the angular velocity required to stabilize the image motion. We demonstrate that, under certain assumptions, the compensatory rotational flow is inversely proportional to the object's distance. The proposed approach achieves new state-of-the-art accuracy in distance estimation - a performance gain of 16% on EVIMO2. EVIMO2 event sequences comprise complex camera motion and substantial variance in depth of static real world scenes.

Recently, research on open domain dialogue systems have attracted extensive interests of academic and industrial researchers. The goal of an open domain dialogue system is to imitate humans in conversations. Previous works on single turn conversation generation have greatly promoted the research of open domain dialogue systems. However, understanding multiple single turn conversations is not equal to the understanding of multi turn dialogue due to the coherent and context dependent properties of human dialogue. Therefore, in open domain multi turn dialogue generation, it is essential to modeling the contextual semantics of the dialogue history, rather than only according to the last utterance. Previous research had verified the effectiveness of the hierarchical recurrent encoder-decoder framework on open domain multi turn dialogue generation. However, using RNN-based model to hierarchically encoding the utterances to obtain the representation of dialogue history still face the problem of a vanishing gradient. To address this issue, in this paper, we proposed a static and dynamic attention-based approach to model the dialogue history and then generate open domain multi turn dialogue responses. Experimental results on Ubuntu and Opensubtitles datasets verify the effectiveness of the proposed static and dynamic attention-based approach on automatic and human evaluation metrics in various experimental settings. Meanwhile, we also empirically verify the performance of combining the static and dynamic attentions on open domain multi turn dialogue generation., Comment: published as a journal paper at ACM Transactions on Information Systems 2023. 30 pages, 6 figures

Large language models (LLMs) have achieved high accuracy, i.e., more than 90% pass@1, in solving Python coding problems in HumanEval and MBPP. Thus, a natural question is, whether LLMs achieve comparable code completion performance compared to human developers? Unfortunately, one cannot answer this question using existing manual crafted or simple (e.g., single-line) code generation benchmarks, since such tasks fail to represent real-world software development tasks. In addition, existing benchmarks often use poor code correctness metrics, providing misleading conclusions. To address these challenges, we create REPOCOD, a code generation benchmark with 980 problems collected from 11 popular real-world projects, with more than 58% of them requiring file-level or repository-level context information. In addition, REPOCOD has the longest average canonical solution length (331.6 tokens) and the highest average cyclomatic complexity (9.00) compared to existing benchmarks. Each task in REPOCOD includes 313.5 developer-written test cases on average for better correctness evaluation. In our evaluations of ten LLMs, none of the models achieve more than 30% pass@1 on REPOCOD, indicating the necessity of building stronger LLMs that can help developers in real-world software development. REPOCOD is available at https://github.com/lt-asset/REPOCOD

We report the discovery of a peculiar X-ray transient, EP240408a, by Einstein Probe (EP) and follow-up studies made with EP, Swift, NICER, GROND, ATCA and other ground-based multi-wavelength telescopes. The new transient was first detected with Wide-field X-ray Telescope (WXT) on board EP on April 8th, 2024, manifested in an intense yet brief X-ray flare lasting for 12 seconds. The flare reached a peak flux of 3.9x10^(-9) erg/cm2/s in 0.5-4 keV, about 300 times brighter than the underlying X-ray emission detected throughout the observation. Rapid and more precise follow-up observations by EP/FXT, Swift and NICER confirmed the finding of this new transient. Its X-ray spectrum is non-thermal in 0.5-10 keV, with a power-law photon index varying within 1.8-2.5. The X-ray light curve shows a plateau lasting for about 4 days, followed by a steep decay till becoming undetectable about 10 days after the initial detection. Based on its temporal property and constraints from previous EP observations, an unusual timescale in the range of 7-23 days is found for EP240408a, which is intermediate between the commonly found fast and long-term transients. No counterparts have been found in optical and near-infrared, with the earliest observation at 17 hours after the initial X-ray detection, suggestive of intrinsically weak emission in these bands. We demonstrate that the remarkable properties of EP240408a are inconsistent with any of the transient types known so far, by comparison with, in particular, jetted tidal disruption events, gamma-ray bursts, X-ray binaries and fast blue optical transients. The nature of EP240408a thus remains an enigma. We suggest that EP240408a may represent a new type of transients with intermediate timescales of the order of about 10 days. The detection and follow-ups of more of such objects are essential for revealing their origin., Comment: 25 pages, 11 figures

Gravitational waves (GWs) from stellar binary black hole (sBBH) mergers can be strongly gravitational lensed by intervening galaxies/galaxy clusters. Only a few works investigated the cluster-lensed sBBH mergers by adopting oversimplified models, while galaxy-lensed ones were intensively studied. In this paper, we estimate the detection rate of cluuster-lensed sBBH mergers with the third-generation GW detectors and its dependence on the lens models. We adopt detailed modeling of galaxy cluster lenses by using the mock clusters in the Synthetic Sky Catalog for Dark Energy Science with LSST (CosmoDC2) and/or approximations of the pseudo-Jaffe profile or an eccentric Navarro-Frenk-White dark matter halo plus a bright central galaxy with singular isothermal sphere profile. Considering the formation of sBBH mergers dominates by the channel of evolution of massive binary stars (EMBS), we find that the detection rate of cluster-lensed sBBHs is $\sim5-84$ yr$^{-1}$, depending on the adopted lens model and uncertainty in the merger rate density, and it is about $\sim{13_{-2.0}^{+28}}$yr$^{-1}$ if adopting relatively more realistic galaxy clusters with central main and member galaxies in the CosmoDC2 catalog, close to the estimated detection rate of sBBH mergers lensed by galaxies. In addition, we also consider the case that the production of sBBH mergers dominated by the dynamical interactions in dense stellar systems. We find that the detection rate of cluster-lensed sBBHs if from the dynamical channel is about $1.5$ times larger than that from the EMBS channel and the redshift distribution of former peaking at a higher redshift ($\sim3$) compared with that from latter ($\sim2$)., Comment: 12 pages, 5 figures, accepted by ApJ

The attention mechanism plays an important role in the machine reading comprehension (MRC) model. Here, we describe a pipeline for building an MRC model with a pretrained language model and visualizing the effect of each attention zone in different layers, which can indicate the explainability of the model. With the presented protocol and accompanying code, researchers can easily visualize the relevance of each attention zone in the MRC model. This approach can be generalized to other pretrained language models., Comment: 17 pages, published in STAR Protocols

Low-rank adaption (LoRA) is a widely used parameter-efficient finetuning method for LLM that reduces memory requirements. However, current LoRA optimizers lack transformation invariance, meaning the actual updates to the weights depends on how the two LoRA factors are scaled or rotated. This deficiency leads to inefficient learning and sub-optimal solutions in practice. This paper introduces LoRA-RITE, a novel adaptive matrix preconditioning method for LoRA optimization, which can achieve transformation invariance and remain computationally efficient. We provide theoretical analysis to demonstrate the benefit of our method and conduct experiments on various LLM tasks with different models including Gemma 2B, 7B, and mT5-XXL. The results demonstrate consistent improvements against existing optimizers. For example, replacing Adam with LoRA-RITE during LoRA fine-tuning of Gemma-2B yielded 4.6\% accuracy gain on Super-Natural Instructions and 3.5\% accuracy gain across other four LLM benchmarks (HellaSwag, ArcChallenge, GSM8K, OpenBookQA).

With the resurgent interest in building open-domain dialogue systems, the dialogue generation task has attracted increasing attention over the past few years. This task is usually formulated as a conditional generation problem, which aims to generate a natural and meaningful response given dialogue contexts and specific constraints, such as persona. And maintaining a consistent persona is essential for the dialogue systems to gain trust from the users. Although tremendous advancements have been brought, traditional persona-based dialogue models are typically trained by leveraging a large number of persona-dense dialogue examples. Yet, such persona-dense training data are expensive to obtain, leading to a limited scale. This work presents a novel approach to learning from limited training examples by regarding consistency understanding as a regularization of response generation. To this end, we propose a novel stack-propagation framework for learning a generation and understanding pipeline.Specifically, the framework stacks a Transformer encoder and two Transformer decoders, where the first decoder models response generation and the second serves as a regularizer and jointly models response generation and consistency understanding. The proposed framework can benefit from the stacked encoder and decoders to learn from much smaller personalized dialogue data while maintaining competitive performance. Under different low-resource settings, subjective and objective evaluations prove that the stack-propagation framework outperforms strong baselines in response quality and persona consistency and largely overcomes the shortcomings of traditional models that rely heavily on the persona-dense dialogue data., Comment: published as a journal paper at ACM Transactions on Information Systems 2023. 35 pages, 5 figures

Deep learning is widely used to predict complex dynamical systems in many scientific and engineering areas. However, the black-box nature of these deep learning models presents significant challenges for carrying out simultaneous data assimilation (DA), which is a crucial technique for state estimation, model identification, and reconstructing missing data. Integrating ensemble-based DA methods with nonlinear deep learning models is computationally expensive and may suffer from large sampling errors. To address these challenges, we introduce a deep learning framework designed to simultaneously provide accurate forecasts and efficient DA. It is named Conditional Gaussian Koopman Network (CGKN), which transforms general nonlinear systems into nonlinear neural differential equations with conditional Gaussian structures. CGKN aims to retain essential nonlinear components while applying systematic and minimal simplifications to facilitate the development of analytic formulae for nonlinear DA. This allows for seamless integration of DA performance into the deep learning training process, eliminating the need for empirical tuning as required in ensemble methods. CGKN compensates for structural simplifications by lifting the dimension of the system, which is motivated by Koopman theory. Nevertheless, CGKN exploits special nonlinear dynamics within the lifted space. This enables the model to capture extreme events and strong non-Gaussian features in joint and marginal distributions with appropriate uncertainty quantification. We demonstrate the effectiveness of CGKN for both prediction and DA on three strongly nonlinear and non-Gaussian turbulent systems: the projected stochastic Burgers--Sivashinsky equation, the Lorenz 96 system, and the El Ni\~no-Southern Oscillation. The results justify the robustness and computational efficiency of CGKN.

In recent years, the rapid aging of the global population has led to an increase in cognitive disorders, such as Alzheimer's disease, presenting significant public health challenges. Although no effective treatments currently exist to reverse Alzheimer's, prevention and early intervention, including cognitive training, are critical. This report explores the potential of AI chatbots in enhancing personalized cognitive training. We introduce ReMe, a web-based framework designed to create AI chatbots that facilitate cognitive training research, specifically targeting episodic memory tasks derived from personal life logs. By leveraging large language models, ReMe provides enhanced user-friendly, interactive, and personalized training experiences. Case studies demonstrate ReMe's effectiveness in engaging users through life recall and open-ended language puzzles, highlighting its potential to improve cognitive training design. Despite promising results, further research is needed to validate training effectiveness through large-scale studies that include cognitive ability evaluations. Overall, ReMe offers a promising approach to personalized cognitive training, utilizing AI capabilities to meet the growing demand for non-pharmacological interventions in cognitive health, with future research aiming to expand its applications and efficacy.

Classifier-Free Guidance (CFG), which combines the conditional and unconditional score functions with two coefficients summing to one, serves as a practical technique for diffusion model sampling. Theoretically, however, denoising with CFG cannot be expressed as a reciprocal diffusion process, which may consequently leave some hidden risks during use. In this work, we revisit the theory behind CFG and rigorously confirm that the improper configuration of the combination coefficients (i.e., the widely used summing-to-one version) brings about expectation shift of the generative distribution. To rectify this issue, we propose ReCFG with a relaxation on the guidance coefficients such that denoising with ReCFG strictly aligns with the diffusion theory. We further show that our approach enjoys a closed-form solution given the guidance strength. That way, the rectified coefficients can be readily pre-computed via traversing the observed data, leaving the sampling speed barely affected. Empirical evidence on real-world data demonstrate the compatibility of our post-hoc design with existing state-of-the-art diffusion models, including both class-conditioned ones (e.g., EDM2 on ImageNet) and text-conditioned ones (e.g., SD3 on CC12M), without any retraining. We will open-source the code to facilitate further research.