Search

Large language models have shown significant capabilities across various domains, including symbolic music generation. However, leveraging these pre-trained models for controllable music arrangement tasks, each requiring different forms of musical information as control, remains a novel challenge. In this paper, we propose a unified sequence-to-sequence framework that enables the fine-tuning of a symbolic music language model for multiple multi-track arrangement tasks, including band arrangement, piano reduction, drum arrangement, and voice separation. Our experiments demonstrate that the proposed approach consistently achieves higher musical quality compared to task-specific baselines across all four tasks. Furthermore, through additional experiments on probing analysis, we show the pre-training phase equips the model with essential knowledge to understand musical conditions, which is hard to acquired solely through task-specific fine-tuning., Comment: Submitted to AAAI 2025

Heterogeneous Graph Neural Networks (HGNNs) have expanded graph representation learning to heterogeneous graph fields. Recent studies have demonstrated their superior performance across various applications, including medical analysis and recommendation systems, often surpassing existing methods. However, GPUs often experience inefficiencies when executing HGNNs due to their unique and complex execution patterns. Compared to traditional Graph Neural Networks, these patterns further exacerbate irregularities in memory access. To tackle these challenges, recent studies have focused on developing domain-specific accelerators for HGNNs. Nonetheless, most of these efforts have concentrated on optimizing the datapath or scheduling data accesses, while largely overlooking the potential benefits that could be gained from leveraging the inherent properties of the semantic graph, such as its topology, layout, and generation. In this work, we focus on leveraging the properties of semantic graphs to enhance HGNN performance. First, we analyze the Semantic Graph Build (SGB) stage and identify significant opportunities for data reuse during semantic graph generation. Next, we uncover the phenomenon of buffer thrashing during the Graph Feature Processing (GFP) stage, revealing potential optimization opportunities in semantic graph layout. Furthermore, we propose a lightweight hardware accelerator frontend for HGNNs, called SiHGNN. This accelerator frontend incorporates a tree-based Semantic Graph Builder for efficient semantic graph generation and features a novel Graph Restructurer for optimizing semantic graph layouts. Experimental results show that SiHGNN enables the state-of-the-art HGNN accelerator to achieve an average performance improvement of 2.95$\times$., Comment: 12 pages, 18 figures. arXiv admin note: text overlap with arXiv:2404.04792

Pulsar glitch is a phenomenon characterized by abrupt changes in the spin period over less than a minute. We present a comprehensive analysis of glitches in four gamma-ray pulsars by combining the timing observation data of \textit{Fermi} Large Area Telescope (\textit{Fermi}-LAT) and Parkes 64 m radio telescope. The timing data of five pulsars, namely PSRs J1028$-$5819, J1420$-$6048, J1509$-$5850, J1709$-$4429 (B1706$-$44) and J1718$-$3825, spanning over 14 years of observations for each, are examined. A total of 12 glitches are identified in four pulsars, including a previously unreported glitch. That is, a new small glitch is identified for PSR J1718$-$3825 in MJD $\sim$ 59121(8), and the fractional glitch size was $\Delta \nu/\nu \sim 1.9(2) \times 10^{-9}$. For PSR J1420$-$6048, our investigation confirms the existence of two linear recovery terms during the evolution of $\dot{\nu}$ subsequent to glitches 4, 6 and 8, and identified an exponential recovery process in glitch 8, with $Q = 0.0131(5)$, $\tau_{\rm d} = 100(6)$ d. Regarding the fourth glitch of PSR J1709$-$4429, our analysis reveals the presence of two exponential recovery terms with healing parameters and decay time-scales $Q$1=0.0104(5), $\tau_{\rm d1}=72(4)$ d and $Q$2 = 0.006(1), $\tau_{\rm d2}=4.2(6)$ d, respectively. For the remaining previously reported glitches, we refine the glitch epochs and glitch observables through precise fitting of the timing residual data. We extensively discuss how multi-band data of glitches can help better characterize the glitch recoveries and constrain the underlying physics of glitch events. We demonstrate that the accumulation of observational data reveals the rich complexity of the glitch phenomenon, aiding in the search for a well-established interpretation., Comment: 16 pages, 8 figures, 7 tables

Parameter-efficient fine-tuning (PEFT) techniques have emerged to address issues of overfitting and high computational costs associated with fully fine-tuning in the paradigm of self-supervised learning. Mainstream methods based on PEFT involve adding a few trainable parameters while keeping the pre-trained parameters of the backbone fixed. These methods achieve comparative, and often superior, performance to fully fine-tuning, demonstrating the powerful representation ability of the pre-trained backbone. Despite its success, these methods typically ignore the initialization of the new parameters, often relying solely on random initialization. We argue that if pre-training is significantly beneficial, it should be applied to all parameters requiring representational capacity. Motivated by this insight, we propose a simple yet effective fine-tuning framework based on Target Parameter Pre-training (TPP). The target parameters refer to the new parameters introduced during fine-tuning. TPP includes an additional stage before PEFT to pre-train these target parameters. During this stage, the pre-trained backbone parameters are frozen, and only the target parameters are trainable. A defined pre-text task is used to encourage the target parameters to learn specific representations of downstream data. When PEFT is subsequently employed, the pre-trained target parameters are loaded to enhance fine-tuning efficiency. The proposed TPP framework is versatile, allowing for the integration of various pretext tasks for pre-training and supporting different PEFT methods as backbones. We evaluated the fine-tining performance of our method using five public datasets, including three modalities and two task types. The results demonstrate that the proposed TPP can be easily integrated into existing PEFT methods, significantly improving performance.

Language Language Models (LLMs) face safety concerns due to potential misuse by malicious users. Recent red-teaming efforts have identified adversarial suffixes capable of jailbreaking LLMs using the gradient-based search algorithm Greedy Coordinate Gradient (GCG). However, GCG struggles with computational inefficiency, limiting further investigations regarding suffix transferability and scalability across models and data. In this work, we bridge the connection between search efficiency and suffix transferability. We propose a two-stage transfer learning framework, DeGCG, which decouples the search process into behavior-agnostic pre-searching and behavior-relevant post-searching. Specifically, we employ direct first target token optimization in pre-searching to facilitate the search process. We apply our approach to cross-model, cross-data, and self-transfer scenarios. Furthermore, we introduce an interleaved variant of our approach, i-DeGCG, which iteratively leverages self-transferability to accelerate the search process. Experiments on HarmBench demonstrate the efficiency of our approach across various models and domains. Notably, our i-DeGCG outperforms the baseline on Llama2-chat-7b with ASRs of $43.9$ ($+22.2$) and $39.0$ ($+19.5$) on valid and test sets, respectively. Further analysis on cross-model transfer indicates the pivotal role of first target token optimization in leveraging suffix transferability for efficient searching., Comment: 11 pages, 4 figures

Out-of-distribution (OOD) detection, which determines whether a given sample is part of the in-distribution (ID), has recently shown promising results through training with synthetic OOD datasets. Nonetheless, existing methods often produce outliers that are considerably distant from the ID, showing limited efficacy for capturing subtle distinctions between ID and OOD. To address these issues, we propose a novel framework, Semantic Outlier generation via Nuisance Awareness (SONA), which notably produces challenging outliers by directly leveraging pixel-space ID samples through diffusion models. Our approach incorporates SONA guidance, providing separate control over semantic and nuisance regions of ID samples. Thereby, the generated outliers achieve two crucial properties: (i) they present explicit semantic-discrepant information, while (ii) maintaining various levels of nuisance resemblance with ID. Furthermore, the improved OOD detector training with SONA outliers facilitates learning with a focus on semantic distinctions. Extensive experiments demonstrate the effectiveness of our framework, achieving an impressive AUROC of 88% on near-OOD datasets, which surpasses the performance of baseline methods by a significant margin of approximately 6%.

We parameterize the equation of state of late-time dark energy as $w_{\mathrm{bin}}(z)$, with three redshift bins, characterized by a constant equation of state in each bin. Then, we constrain the parameters of the $w_{\mathrm{bin}}$CDM model using datasets from DESI BAO data, Planck CMB power spectrum, ACT DR6 lensing power spectrum, and type Ia supernova distance-redshift data of Pantheon Plus/DES Y5/Union3. The significances for $w_1>-1$ is $1.9\sigma$, $2.6\sigma$ and $3.3\sigma$, and $w_2$ is consistent with $-1$ within $1\sigma$ level, while $1.6\sigma$, $1.5\sigma$ and $1.5\sigma$ for $w_3<-1$ in these three data combinations with different choices of type Ia supernova datasets, respectively. Additionally, to alleviate $H_0$ tension, we incorporate the early dark energy (EDE) model in the early-time universe and add the SH0ES absolute magnitude $M_b$ prior (or $H_0$ prior) to further constrain the $w_{\mathrm{bin}}$EDE model. In the $w_{\mathrm{bin}}$EDE model, we find a $w_{\mathrm{bin}}(z)$ pattern similar to that in the $w_{\mathrm{bin}}$CDM model. The results of the three data combinations exhibit $w_1>-1$ at $1.9\sigma$, $1.7\sigma$ and $2.9\sigma$ level, meanwhile $w_3<-1$ at $1.3\sigma$, $1.3\sigma$ and $1.3\sigma$ level, respectively. In all, our results indicate that the transition of dark energy from phantom at high redshifts to quintessence at low redshifts is not conclusive., Comment: 10 pages, 4 figures, 2 tables

The advent of large language models such as ChatGPT, Gemini, and others has underscored the importance of evaluating their diverse capabilities, ranging from natural language understanding to code generation. However, their performance on spatial tasks has not been comprehensively assessed. This study addresses this gap by introducing a novel multi-task spatial evaluation dataset, designed to systematically explore and compare the performance of several advanced models on spatial tasks. The dataset encompasses twelve distinct task types, including spatial understanding and path planning, each with verified, accurate answers. We evaluated multiple models, including OpenAI's gpt-3.5-turbo, gpt-4o, and ZhipuAI's glm-4, through a two-phase testing approach. Initially, we conducted zero-shot testing, followed by categorizing the dataset by difficulty and performing prompt tuning tests. Results indicate that gpt-4o achieved the highest overall accuracy in the first phase, with an average of 71.3%. Although moonshot-v1-8k slightly underperformed overall, it surpassed gpt-4o in place name recognition tasks. The study also highlights the impact of prompt strategies on model performance in specific tasks. For example, the Chain-of-Thought (COT) strategy increased gpt-4o's accuracy in path planning from 12.4% to 87.5%, while a one-shot strategy enhanced moonshot-v1-8k's accuracy in mapping tasks from 10.1% to 76.3%.

The Metaverse, a burgeoning collective virtual space merging augmented reality and persistent virtual worlds, necessitates advanced artificial intelligence (AI) and communication technologies to support immersive and interactive experiences. Federated learning (FL) has emerged as a promising technique for collaboratively training AI models while preserving data privacy. However, FL faces challenges such as high communication overhead and substantial computational demands, particularly for neural network (NN) models. To address these issues, we propose an integrated federated split learning and hyperdimensional computing (FSL-HDC) framework for emerging foundation models. This novel approach reduces communication costs, computation load, and privacy risks, making it particularly suitable for resource-constrained edge devices in the Metaverse, ensuring real-time responsive interactions. Additionally, we introduce an optimization algorithm that concurrently optimizes transmission power and bandwidth to minimize the maximum transmission time among all users to the server. The simulation results based on the MNIST dataset indicate that FSL-HDC achieves an accuracy rate of approximately 87.5%, which is slightly lower than that of FL-HDC. However, FSL-HDC exhibits a significantly faster convergence speed, approximately 3.733x that of FSL-NN, and demonstrates robustness to non-IID data distributions. Moreover, our proposed optimization algorithm can reduce the maximum transmission time by up to 64% compared with the baseline.

In this study, we use the maximum likelihood estimator (MLE) to explore factorization and event-plane correlations in relativistic heavy-ion collisions. Our analyses incorporate both numerical simulations and publicly available data from the CMS Collaboration. We focus on Au+Au collisions at 200 GeV and Pb+Pb collisions at 2.76 TeV. The differential flows obtained for various centrality windows and momentum cuts are consistent with conventional methodologies such as multi-particle cumulants and event-plane methods. Leveraging these findings, we proceed to undertake further analysis of flow factorization and event-plane correlations. These quantities are relevant because of their sensitivity to initial-state fluctuations. While higher-order correlators might provide different implementations of factorization ratio, the MLE estimator is readily applied to these scenarios. Moreover, MLE's unique capabilities allow us to compute specific correlators that are typically inaccessible by other means. As an asymptotically normal and unbiased estimator, MLE provides a valuable alternative tool for flow and correlation analysis., Comment: 21 pages and 12 figures

Existing works in single-image human reconstruction suffer from weak generalizability due to insufficient training data or 3D inconsistencies for a lack of comprehensive multi-view knowledge. In this paper, we introduce MagicMan, a human-specific multi-view diffusion model designed to generate high-quality novel view images from a single reference image. As its core, we leverage a pre-trained 2D diffusion model as the generative prior for generalizability, with the parametric SMPL-X model as the 3D body prior to promote 3D awareness. To tackle the critical challenge of maintaining consistency while achieving dense multi-view generation for improved 3D human reconstruction, we first introduce hybrid multi-view attention to facilitate both efficient and thorough information interchange across different views. Additionally, we present a geometry-aware dual branch to perform concurrent generation in both RGB and normal domains, further enhancing consistency via geometry cues. Last but not least, to address ill-shaped issues arising from inaccurate SMPL-X estimation that conflicts with the reference image, we propose a novel iterative refinement strategy, which progressively optimizes SMPL-X accuracy while enhancing the quality and consistency of the generated multi-views. Extensive experimental results demonstrate that our method significantly outperforms existing approaches in both novel view synthesis and subsequent 3D human reconstruction tasks., Comment: Project Page: https://thuhcsi.github.io/MagicMan

Significant progress has been made in the field of Instruction-based Image Editing (IIE). However, evaluating these models poses a significant challenge. A crucial requirement in this field is the establishment of a comprehensive evaluation benchmark for accurately assessing editing results and providing valuable insights for its further development. In response to this need, we propose I2EBench, a comprehensive benchmark designed to automatically evaluate the quality of edited images produced by IIE models from multiple dimensions. I2EBench consists of 2,000+ images for editing, along with 4,000+ corresponding original and diverse instructions. It offers three distinctive characteristics: 1) Comprehensive Evaluation Dimensions: I2EBench comprises 16 evaluation dimensions that cover both high-level and low-level aspects, providing a comprehensive assessment of each IIE model. 2) Human Perception Alignment: To ensure the alignment of our benchmark with human perception, we conducted an extensive user study for each evaluation dimension. 3) Valuable Research Insights: By analyzing the advantages and disadvantages of existing IIE models across the 16 dimensions, we offer valuable research insights to guide future development in the field. We will open-source I2EBench, including all instructions, input images, human annotations, edited images from all evaluated methods, and a simple script for evaluating the results from new IIE models. The code, dataset and generated images from all IIE models are provided in github: https://github.com/cocoshe/I2EBench., Comment: Tech report, 39 pages, 41 figures

Neural Architecture Search (NAS) has proven effective in discovering new Convolutional Neural Network (CNN) architectures, particularly for scenarios with well-defined accuracy optimization goals. However, previous approaches often involve time-consuming training on super networks or intensive architecture sampling and evaluations. Although various zero-cost proxies correlated with CNN model accuracy have been proposed for efficient architecture search without training, their lack of hardware consideration makes it challenging to target highly resource-constrained edge devices such as microcontroller units (MCUs). To address these challenges, we introduce MONAS, a novel hardware-aware zero-shot NAS framework specifically designed for MCUs in edge computing. MONAS incorporates hardware optimality considerations into the search process through our proposed MCU hardware latency estimation model. By combining this with specialized performance indicators (proxies), MONAS identifies optimal neural architectures without incurring heavy training and evaluation costs, optimizing for both hardware latency and accuracy under resource constraints. MONAS achieves up to a 1104x improvement in search efficiency over previous work targeting MCUs and can discover CNN models with over 3.23x faster inference on MCUs while maintaining similar accuracy compared to more general NAS approaches.

The rapid progress in Deep Learning (DL) and Large Language Models (LLMs) has exponentially increased demands of computational power and bandwidth. This, combined with the high costs of faster computing chips and interconnects, has significantly inflated High Performance Computing (HPC) construction costs. To address these challenges, we introduce the Fire-Flyer AI-HPC architecture, a synergistic hardware-software co-design framework and its best practices. For DL training, we deployed the Fire-Flyer 2 with 10,000 PCIe A100 GPUs, achieved performance approximating the DGX-A100 while reducing costs by half and energy consumption by 40%. We specifically engineered HFReduce to accelerate allreduce communication and implemented numerous measures to keep our Computation-Storage Integrated Network congestion-free. Through our software stack, including HaiScale, 3FS, and HAI-Platform, we achieved substantial scalability by overlapping computation and communication. Our system-oriented experience from DL training provides valuable insights to drive future advancements in AI-HPC., Comment: This is the preprint version of the paper accepted for presentation at the 2024 International Conference for High Performance Computing, Networking, Storage, and Analysis (SC'24). \c{opyright} 2024 IEEE. Personal use of this material is permitted. For other uses, permission from IEEE must be obtained. Please refer to IEEE Xplore for the final published version

Satellite-to-street view synthesis aims at generating a realistic street-view image from its corresponding satellite-view image. Although stable diffusion models have exhibit remarkable performance in a variety of image generation applications, their reliance on similar-view inputs to control the generated structure or texture restricts their application to the challenging cross-view synthesis task. In this work, we propose CrossViewDiff, a cross-view diffusion model for satellite-to-street view synthesis. To address the challenges posed by the large discrepancy across views, we design the satellite scene structure estimation and cross-view texture mapping modules to construct the structural and textural controls for street-view image synthesis. We further design a cross-view control guided denoising process that incorporates the above controls via an enhanced cross-view attention module. To achieve a more comprehensive evaluation of the synthesis results, we additionally design a GPT-based scoring method as a supplement to standard evaluation metrics. We also explore the effect of different data sources (e.g., text, maps, building heights, and multi-temporal satellite imagery) on this task. Results on three public cross-view datasets show that CrossViewDiff outperforms current state-of-the-art on both standard and GPT-based evaluation metrics, generating high-quality street-view panoramas with more realistic structures and textures across rural, suburban, and urban scenes. The code and models of this work will be released at https://opendatalab.github.io/CrossViewDiff/., Comment: 21 pages, 11 figures

Quantum machine learning QML algorithms promise to deliver near-term, applicable quantum computation on noisy, intermediate-scale systems. While most of these algorithms leverage quantum circuits for generic applications, a recent set of proposals, called analog quantum machine learning (AQML) algorithms, breaks away from circuit-based abstractions and favors leveraging the natural dynamics of quantum systems for computation, promising to be noise-resilient and suited for specific applications such as quantum simulation. Recent AQML studies have called for determining best ansatz selection practices and whether AQML algorithms have trap-free landscapes based on theory from quantum optimal control (QOC). We address this call by systematically studying AQML landscapes on two models: those admitting black-boxed expressivity and those tailored to simulating a specific unitary evolution. Numerically, the first kind exhibits local traps in their landscapes, while the second kind is trap-free. However, both kinds violate QOC theory's key assumptions for guaranteeing trap-free landscapes. We propose a methodology to co-design AQML algorithms for unitary evolution simulation using the ansatz's Magnus expansion. We show favorable convergence in simulating dynamics with applications to metrology and quantum chemistry. We conclude that such co-design is necessary to ensure the applicability of AQML algorithms., Comment: 10 pages, 6 figures

This paper introduces a novel framework designed to achieve a high compression ratio in Split Learning (SL) scenarios where resource-constrained devices are involved in large-scale model training. Our investigations demonstrate that compressing feature maps within SL leads to biased gradients that can negatively impact the convergence rates and diminish the generalization capabilities of the resulting models. Our theoretical analysis provides insights into how compression errors critically hinder SL performance, which previous methodologies underestimate. To address these challenges, we employ a narrow bit-width encoded mask to compensate for the sparsification error without increasing the order of time complexity. Supported by rigorous theoretical analysis, our framework significantly reduces compression errors and accelerates the convergence. Extensive experiments also verify that our method outperforms existing solutions regarding training efficiency and communication complexity.

Frequency-time is a degree of freedom suitable for photonic high-dimensional entanglement, with advantages such as compatibility with single-mode devices and insensitivity to dispersion. The engineering control of the frequency-time amplitude of a photon's electric field has been demonstrated on platforms with second-order optical nonlinearity. For integrated photonic platforms with only third-order optical nonlinearity, the engineered generation of the state remains unexplored. Here, we demonstrate a cavity-enhanced photon-pair source on the silicon-on-insulator (SOI) platform that can generate both separable states and controllable entangled states in the frequency domain without post-manipulation. By choosing different resonance combinations and employing on-chip optical field differentiation, we achieve independent control over two functions that affect the joint spectral intensity (JSI) of the state. A semi-analytical model is derived to simulate the biphoton spectral wavefunction in the presence of resonance splitting and pump differentiation, and its parameters can be fully determined through fitting-based parameter extraction from the resonator's measured linear response. The measured spectral purity for the separable state is $95.5\pm 1.2\%$, while the measured JSIs for the entangled states show two- or four-peaked functions in two-dimensional frequency space. The experiments and simulations demonstrate the capacity to manipulate the frequency-domain wavefunction in a silicon-based device, which is promising for applications like quantum information processing using pulsed temporal-mode encoding or long-distance quantum key distribution.

In a conventional Federated Learning framework, client selection for training typically involves the random sampling of a subset of clients in each iteration. However, this random selection often leads to disparate performance among clients, raising concerns regarding fairness, particularly in applications where equitable outcomes are crucial, such as in medical or financial machine learning tasks. This disparity typically becomes more pronounced with the advent of performance-centric client sampling techniques. This paper introduces two novel methods, namely SUBTRUNC and UNIONFL, designed to address the limitations of random client selection. Both approaches utilize submodular function maximization to achieve more balanced models. By modifying the facility location problem, they aim to mitigate the fairness concerns associated with random selection. SUBTRUNC leverages client loss information to diversify solutions, while UNIONFL relies on historical client selection data to ensure a more equitable performance of the final model. Moreover, these algorithms are accompanied by robust theoretical guarantees regarding convergence under reasonable assumptions. The efficacy of these methods is demonstrated through extensive evaluations across heterogeneous scenarios, revealing significant improvements in fairness as measured by a client dissimilarity metric., Comment: 13 pages

Videos showcasing specific products are increasingly important for E-commerce. Key moments naturally exist as the first appearance of a specific product, presentation of its distinctive features, the presence of a buying link, etc. Adding proper sound effects (SFX) to these key moments, or video decoration with SFX (VDSFX), is crucial for enhancing the user engaging experience. Previous studies about adding SFX to videos perform video to SFX matching at a holistic level, lacking the ability of adding SFX to a specific moment. Meanwhile, previous studies on video highlight detection or video moment retrieval consider only moment localization, leaving moment to SFX matching untouched. By contrast, we propose in this paper D&M, a unified method that accomplishes key moment detection and moment to SFX matching simultaneously. Moreover, for the new VDSFX task we build a large-scale dataset SFX-Moment from an E-commerce platform. For a fair comparison, we build competitive baselines by extending a number of current video moment detection methods to the new task. Extensive experiments on SFX-Moment show the superior performance of the proposed method over the baselines. Code and data will be released., Comment: 9 pages, 4 figures

Healthcare alert systems (HAS) are undergoing rapid evolution, propelled by advancements in artificial intelligence (AI), Internet of Things (IoT) technologies, and increasing health consciousness. Despite significant progress, a fundamental challenge remains: balancing the accuracy of personalized health alerts with stringent privacy protection in HAS environments constrained by resources. To address this issue, we introduce a uniform framework, LLM-HAS, which incorporates Large Language Models (LLM) into HAS to significantly boost the accuracy, ensure user privacy, and enhance personalized health service, while also improving the subjective quality of experience (QoE) for users. Our innovative framework leverages a Mixture of Experts (MoE) approach, augmented with LLM, to analyze users' personalized preferences and potential health risks from additional textual job descriptions. This analysis guides the selection of specialized Deep Reinforcement Learning (DDPG) experts, tasked with making precise health alerts. Moreover, LLM-HAS can process Conversational User Feedback, which not only allows fine-tuning of DDPG but also deepen user engagement, thereby enhancing both the accuracy and personalization of health management strategies. Simulation results validate the effectiveness of the LLM-HAS framework, highlighting its potential as a groundbreaking approach for employing generative AI (GAI) to provide highly accurate and reliable alerts.

The expanding use of Unmanned Aerial Vehicles (UAVs) in vital areas like traffic management, surveillance, and environmental monitoring highlights the need for robust communication and navigation systems. Particularly vulnerable are Global Navigation Satellite Systems (GNSS), which face a spectrum of interference and jamming threats that can significantly undermine their performance. While traditional deep learning approaches are adept at mitigating these issues, they often fall short for UAV applications due to significant computational demands and the complexities of managing large, centralized datasets. In response, this paper introduces Federated Reservoir Computing (FedRC) as a potent and efficient solution tailored to enhance interference classification in GNSS systems used by UAVs. Our experimental results demonstrate that FedRC not only achieves faster convergence but also sustains lower loss levels than traditional models, highlighting its exceptional adaptability and operational efficiency.

Image segmentation is a long-standing challenge in computer vision, studied continuously over several decades, as evidenced by seminal algorithms such as N-Cut, FCN, and MaskFormer. With the advent of foundation models (FMs), contemporary segmentation methodologies have embarked on a new epoch by either adapting FMs (e.g., CLIP, Stable Diffusion, DINO) for image segmentation or developing dedicated segmentation foundation models (e.g., SAM). These approaches not only deliver superior segmentation performance, but also herald newfound segmentation capabilities previously unseen in deep learning context. However, current research in image segmentation lacks a detailed analysis of distinct characteristics, challenges, and solutions associated with these advancements. This survey seeks to fill this gap by providing a thorough review of cutting-edge research centered around FM-driven image segmentation. We investigate two basic lines of research -- generic image segmentation (i.e., semantic segmentation, instance segmentation, panoptic segmentation), and promptable image segmentation (i.e., interactive segmentation, referring segmentation, few-shot segmentation) -- by delineating their respective task settings, background concepts, and key challenges. Furthermore, we provide insights into the emergence of segmentation knowledge from FMs like CLIP, Stable Diffusion, and DINO. An exhaustive overview of over 300 segmentation approaches is provided to encapsulate the breadth of current research efforts. Subsequently, we engage in a discussion of open issues and potential avenues for future research. We envisage that this fresh, comprehensive, and systematic survey catalyzes the evolution of advanced image segmentation systems., Comment: A comprehensive survey of image segmentation in foundation model era (work in progress)

Neural implicit reconstruction via volume rendering has demonstrated its effectiveness in recovering dense 3D surfaces. However, it is non-trivial to simultaneously recover meticulous geometry and preserve smoothness across regions with differing characteristics. To address this issue, previous methods typically employ geometric priors, which are often constrained by the performance of the prior models. In this paper, we propose ND-SDF, which learns a Normal Ddeflection field to represent the angular deviation between the scene normal and the prior normal. Unlike previous methods that uniformly apply geometric priors on all samples, introducing significant bias in accuracy, our proposed normal deflection field dynamically learns and adapts the utilization of samples based on their specific characteristics, thereby improving both the accuracy and effectiveness of the model. Our method not only obtains smooth weakly textured regions such as walls and floors but also preserves the geometric details of complex structures. In addition, we introduce a novel ray sampling strategy based on the deflection angle to facilitate the unbiased rendering process, which significantly improves the quality and accuracy of intricate surfaces, especially on thin structures. Consistent improvements on various challenging datasets demonstrate the superiority of our method.

Understanding people's social interactions in complex real-world scenarios often relies on intricate mental reasoning. To truly understand how and why people interact with one another, we must infer the underlying mental states that give rise to the social interactions, i.e., Theory of Mind reasoning in multi-agent interactions. Additionally, social interactions are often multi-modal -- we can watch people's actions, hear their conversations, and/or read about their past behaviors. For AI systems to successfully and safely interact with people in real-world environments, they also need to understand people's mental states as well as their inferences about each other's mental states based on multi-modal information about their interactions. For this, we introduce MuMA-ToM, a Multi-modal Multi-Agent Theory of Mind benchmark. MuMA-ToM is the first multi-modal Theory of Mind benchmark that evaluates mental reasoning in embodied multi-agent interactions. In MuMA-ToM, we provide video and text descriptions of people's multi-modal behavior in realistic household environments. Based on the context, we then ask questions about people's goals, beliefs, and beliefs about others' goals. We validated MuMA-ToM in a human experiment and provided a human baseline. We also proposed a novel multi-modal, multi-agent ToM model, LIMP (Language model-based Inverse Multi-agent Planning). Our experimental results show that LIMP significantly outperforms state-of-the-art methods, including large multi-modal models (e.g., GPT-4o, Gemini-1.5 Pro) and a recent multi-modal ToM model, BIP-ALM., Comment: Project website: https://scai.cs.jhu.edu/projects/MuMA-ToM/ Code: https://github.com/SCAI-JHU/MuMA-ToM

The Mantel-Haenszel (MH) risk difference estimator, commonly used in randomized clinical trials for binary outcomes, calculates a weighted average of stratum-specific risk difference estimators. Traditionally, this method requires the stringent assumption that risk differences are homogeneous across strata, also known as the common risk difference assumption. In our article, we relax this assumption and adopt a modern perspective, viewing the MH risk difference estimator as an approach for covariate adjustment in randomized clinical trials, distinguishing its use from that in meta-analysis and observational studies. We demonstrate that the MH risk difference estimator consistently estimates the average treatment effect within a standard super-population framework, which is often the primary interest in randomized clinical trials, in addition to estimating a weighted average of stratum-specific risk difference. We rigorously study its properties under both the large-stratum and sparse-stratum asymptotic regimes. Furthermore, for either estimand, we propose a unified robust variance estimator that improves over the popular variance estimators by Greenland and Robins (1985) and Sato et al. (1989) and has provable consistency across both asymptotic regimes, regardless of assuming common risk differences. Extensions of our theoretical results also provide new insights into the Cochran-Mantel-Haenszel test and the post-stratification estimator. Our findings are thoroughly validated through simulations and a clinical trial example.

Medical image datasets in the real world are often unlabeled and imbalanced, and Semi-Supervised Object Detection (SSOD) can utilize unlabeled data to improve an object detector. However, existing approaches predominantly assumed that the unlabeled data and test data do not contain out-of-distribution (OOD) classes. The few open-set semi-supervised object detection methods have two weaknesses: first, the class imbalance is not considered; second, the OOD instances are distinguished and simply discarded during pseudo-labeling. In this paper, we consider the open-set semi-supervised object detection problem which leverages unlabeled data that contain OOD classes to improve object detection for medical images. Our study incorporates two key innovations: Category Control Embed (CCE) and out-of-distribution Detection Fusion Classifier (OODFC). CCE is designed to tackle dataset imbalance by constructing a Foreground information Library, while OODFC tackles open-set challenges by integrating the ``unknown'' information into basic pseudo-labels. Our method outperforms the state-of-the-art SSOD performance, achieving a 4.25 mAP improvement on the public Parasite dataset.

This paper presents MaVEn, an innovative Multi-granularity Visual Encoding framework designed to enhance the capabilities of Multimodal Large Language Models (MLLMs) in multi-image reasoning. Current MLLMs primarily focus on single-image visual understanding, limiting their ability to interpret and integrate information across multiple images. MaVEn addresses this limitation by combining discrete visual symbol sequences, which abstract coarse-grained semantic concepts, with traditional continuous representation sequences that model fine-grained features. This dual approach bridges the semantic gap between visual and textual data, thereby improving the model's ability to process and interpret information from multiple images effectively. Additionally, we design a dynamic reduction mechanism by for long-sequence continuous features to enhance multi-image processing efficiency. Experimental results demonstrate that MaVEn significantly enhances MLLMs' understanding in complex multi-image scenarios, while also improving performance in single-image contexts.

Let $\mathbb{K}$ be an algebraically closed field of characteristic $p>5$. We show the existence of minimal models for pseudo-effective NQC klt generalized pairs in dimension three over $\mathbb{K}$. As a consequence, we prove the termination of flips for pseudo-effective threefold NQC klt generalized pairs over $\mathbb{K}$. This provides a new proof on the termination of flips for pseudo-effective pairs over $\mathbb{K}$ without using the non-vanishing theorems. A key ingredient of our proof is the ACC for lc thresholds in dimension $\leq 3$ and the global ACC in dimension $\leq 2$ for generalized pairs over $\mathbb{K}$., Comment: 31 pages

We systematically compute five-point correlators of chiral primary operators with arbitrary Kaluza-Klein charges at tree-level in $\mathrm{AdS}_5\times\mathrm{S}^3$, and obtain a unified formula. This result serves as the first concrete confirmation for the existence of the hidden eight-dimensional symmetries at the level of five points., Comment: 5 pages, 2 figures and 3 appendices + ancillary files for the results

Large language models (LLMs) are aligned with human preferences by reinforcement learning from human feedback (RLHF). Effective data sampling is crucial for RLHF, as it determines the efficiency of model training, ensuring that models learn from the informative samples. To achieve better data generation, we propose a new sampling method called Preference-Guided Reflective Sampling (PRS). PRS frames the response generation as an optimization process to the explicitly specified user preference described in natural language. It employs a tree-based generation framework to enable an efficient sampling process, which guides the direction of generation through preference and better explores the sampling space with adaptive self-refinement. Notably, PRS can align LLMs to diverse preferences. We study preference-controlled text generation for instruction following and keyword-focused document summarization. Our findings indicate that PRS, across different LLM policies, generates training data with much higher rewards than strong baselines. PRS also excels in post-RL training.

By utilizing statistical methods to analyze bibliographic data, bibliometrics faces inherent limitations in identifying the most significant science and technology achievements and researchers. To overcome this challenge, we present an evaluatology-based science and technology evaluation methodology. At the heart of this approach lies the concept of an extended evaluation condition, encompassing eight crucial components derived from a field. We define four relationships that illustrate the connections among various achievements based on their mapped extended EC components, as well as their temporal and citation links. Within a relationship under an extended evaluation condition, evaluators can effectively compare these achievements by carefully addressing the influence of confounding variables. We establish a real-world evaluation system encompassing an entire collection of achievements, each of which is mapped to several components of an extended EC. Within a specific field like chip technology or open source, we construct a perfect evaluation model that can accurately trace the evolution and development of all achievements in terms of four relationships based on the real-world evaluation system. Building upon the foundation of the perfect evaluation model, we put forth four-round rules to eliminate non-significant achievements by utilizing four relationships. This process allows us to establish a pragmatic evaluation model that effectively captures the essential achievements, serving as a curated collection of the top N achievements within a specific field during a specific timeframe. We present a case study on the top 100 Chip achievements which highlights its practical application and efficacy in identifying significant achievements and researchers that otherwise can not be identified by using bibliometrics., Comment: 18 pages, 8 figures, and 2 tables

In this paper, we propose an agent-based model of information spread, grounded on psychological insights on the formation and spread of beliefs. In our model, we consider a network of individuals who share two opposing types of information on a specific topic (e.g., pro- vs. anti-vaccine stances), and the accumulation of evidence supporting either type of information is modelled by means of a drift-diffusion process. After formalising the model, we put forward a campaign of Monte Carlo simulations to identify population-wide behaviours emerging from agents' exposure to different sources of information, investigating the impact of the number and persistence of such sources, and the role of the network structure through which the individuals interact. We find similar emergent behaviours for all network structures considered. When there is a single type of information, the main observed emergent behaviour is consensus. When there are opposing information sources, both consensus or polarisation can result; the latter occurs if the number and persistence of the sources exceeds some threshold values. Importantly, we find the emergent behaviour is mainly influenced by how long the information sources are present for, as opposed to how many sources there are.

Scribble-based weakly supervised segmentation techniques offer comparable performance to fully supervised methods while significantly reducing annotation costs, making them an appealing alternative. Existing methods often rely on auxiliary tasks to enforce semantic consistency and use hard pseudo labels for supervision. However, these methods often overlook the unique requirements of models trained with sparse annotations. Since the model must predict pixel-wise segmentation maps with limited annotations, the ability to handle varying levels of annotation richness is critical. In this paper, we adopt the principle of `from few to more' and propose MaCo, a weakly supervised framework designed for medical image segmentation. MaCo employs masked context modeling (MCM) and continuous pseudo labels (CPL). MCM uses an attention-based masking strategy to disrupt the input image, compelling the model's predictions to remain consistent with those of the original image. CPL converts scribble annotations into continuous pixel-wise labels by applying an exponential decay function to distance maps, resulting in continuous maps that represent the confidence of each pixel belonging to a specific category, rather than using hard pseudo labels. We evaluate MaCo against other weakly supervised methods using three public datasets. The results indicate that MaCo outperforms competing methods across all datasets, setting a new record in weakly supervised medical image segmentation.

We are living in an age of protest. Although we have an excellent understanding of the factors that predict participation in protest, we understand little about the conditions that foster a sustained (versus transient) movement. How do interactions between supporters and authorities combine to influence whether and how people engage (i.e., using conventional or radical tactics)? This paper introduces a novel, theoretically-founded and empirically-informed agent-based model (DIMESim) to address these questions. We model the complex interactions between the psychological attributes of the protester (agents), the authority to whom the protests are targeted, and the environment that allows protesters to coordinate with each other -- over time, and at a population scale. Where an authority is responsive and failure is contested, a modest sized conventional movement endured. Where authorities repeatedly and incontrovertibly fail the movement, the population disengaged from action but evidenced an ongoing commitment to radicalism (latent radicalism)., Comment: Initial submission version of journal paper

The increasing usage of Large Language Models (LLMs) has resulted in a surging demand for planet-scale serving systems, where tens of thousands of GPUs continuously serve hundreds of millions of users. Consequently, throughput (under reasonable latency constraints) has emerged as a key metric that determines serving systems' performance. To boost throughput, various methods of inter-device parallelism (e.g., data, tensor, pipeline) have been explored. However, existing methods do not consider overlapping the utilization of different resources within a single device, leading to underutilization and sub-optimal performance. We propose NanoFlow, a novel serving framework that exploits intra-device parallelism, which overlaps the usage of resources including compute, memory, and network within a single device through operation co-scheduling. To exploit intra-device parallelism, NanoFlow introduces two key innovations: First, NanoFlow splits requests into nano-batches at the granularity of operations, which breaks the dependency of sequential operations in LLM inference and enables overlapping; then, to get benefit from overlapping, NanoFlow uses an operation-level pipeline with execution unit scheduling, which partitions the device's functional units and simultaneously executes different operations in each unit. NanoFlow automates the pipeline setup using a parameter search algorithm, which enables easily porting NanoFlow to different models. We implement NanoFlow on NVIDIA GPUs and evaluate end-to-end serving throughput on several popular models such as LLaMA-2-70B, Mixtral 8x7B, LLaMA-3-8B, etc.. With practical workloads, NanoFlow provides 1.91x throughput boost compared to state-of-the-art serving systems achieving 59% to 72% of optimal throughput across ported models.

Facial expression recognition (FER) is an important research topic in emotional artificial intelligence. In recent decades, researchers have made remarkable progress. However, current FER paradigms face challenges in generalization, lack semantic information aligned with natural language, and struggle to process both images and videos within a unified framework, making their application in multimodal emotion understanding and human-computer interaction difficult. Multimodal Large Language Models (MLLMs) have recently achieved success, offering advantages in addressing these issues and potentially overcoming the limitations of current FER paradigms. However, directly applying pre-trained MLLMs to FER still faces several challenges. Our zero-shot evaluations of existing open-source MLLMs on FER indicate a significant performance gap compared to GPT-4V and current supervised state-of-the-art (SOTA) methods. In this paper, we aim to enhance MLLMs' capabilities in understanding facial expressions. We first generate instruction data for five FER datasets with Gemini. We then propose a novel MLLM, named EMO-LLaMA, which incorporates facial priors from a pretrained facial analysis network to enhance human facial information. Specifically, we design a Face Info Mining module to extract both global and local facial information. Additionally, we utilize a handcrafted prompt to introduce age-gender-race attributes, considering the emotional differences across different human groups. Extensive experiments show that EMO-LLaMA achieves SOTA-comparable or competitive results across both static and dynamic FER datasets. The instruction dataset and code are available at https://github.com/xxtars/EMO-LLaMA.

Large Language Models (LLMs) demonstrate human-level capabilities in dialogue, reasoning, and knowledge retention. However, even the most advanced LLMs face challenges such as hallucinations and real-time updating of their knowledge. Current research addresses this bottleneck by equipping LLMs with external knowledge, a technique known as Retrieval Augmented Generation (RAG). However, two key issues constrained the development of RAG. First, there is a growing lack of comprehensive and fair comparisons between novel RAG algorithms. Second, open-source tools such as LlamaIndex and LangChain employ high-level abstractions, which results in a lack of transparency and limits the ability to develop novel algorithms and evaluation metrics. To close this gap, we introduce RAGLAB, a modular and research-oriented open-source library. RAGLAB reproduces 6 existing algorithms and provides a comprehensive ecosystem for investigating RAG algorithms. Leveraging RAGLAB, we conduct a fair comparison of 6 RAG algorithms across 10 benchmarks. With RAGLAB, researchers can efficiently compare the performance of various algorithms and develop novel algorithms., Comment: 6 pages, 3 figures

Finding optimal control strategies to suppress quantum thermalization for arbitrarily initial states, the so-called quantum nonergodicity control, is important for quantum information science and technologies. Previous control methods largely relied on theoretical model of the target quantum system, but invertible model approximations and inaccuracies can lead to control failures. We develop a model-free and deep-reinforcement learning (DRL) framework for quantum nonergodicity control. It is a machine-learning method with the unique focus on balancing exploration and exploitation strategies to maximize the cumulative rewards so as to preserve the initial memory in the time-dependent nonergodic metrics over a long stretch of time. We use the paradigmatic one-dimensional tilted Fermi-Hubbard system to demonstrate that the DRL agent can efficiently learn the quantum many-body system solely through the interactions with the environment. The optimal policy obtained by the DRL provides broader control scenarios for managing nonergodicity in the phase diagram as compared to, e.g., the specific protocol for Wannier-Stark localization. The continuous control protocols and observations are experimentally feasible. The model-free nature of DRL and its versatile search space for control functions render promising nonergodicity control in more complex quantum many-body systems., Comment: 19 pages, 8 figures

Video Shadow Detection (VSD) aims to detect the shadow masks with frame sequence. Existing works suffer from inefficient temporal learning. Moreover, few works address the VSD problem by considering the characteristic (i.e., boundary) of shadow. Motivated by this, we propose a Timeline and Boundary Guided Diffusion (TBGDiff) network for VSD where we take account of the past-future temporal guidance and boundary information jointly. In detail, we design a Dual Scale Aggregation (DSA) module for better temporal understanding by rethinking the affinity of the long-term and short-term frames for the clipped video. Next, we introduce Shadow Boundary Aware Attention (SBAA) to utilize the edge contexts for capturing the characteristics of shadows. Moreover, we are the first to introduce the Diffusion model for VSD in which we explore a Space-Time Encoded Embedding (STEE) to inject the temporal guidance for Diffusion to conduct shadow detection. Benefiting from these designs, our model can not only capture the temporal information but also the shadow property. Extensive experiments show that the performance of our approach overtakes the state-of-the-art methods, verifying the effectiveness of our components. We release the codes, weights, and results at \url{https://github.com/haipengzhou856/TBGDiff}., Comment: ACM MM2024

Random feature attention (RFA) adopts random fourier feature (RFF) methods to approximate the softmax function, resulting in a linear time and space attention mechanism that enables the construction of an efficient Transformer. Inspired by RFA, we propose Macformer, a Transformer architecture that employs random Maclaurin features (RMF) to approximate various dot-product kernels, thereby accelerating attention computations for long sequence. Macformer consists of Random Maclaurin Feature Attention (RMFA) and pre-post Scaling Batch Normalization (ppSBN), the former is an unbiased approximation for dot-product kernelized attention and the later is a two-stage regularization mechanism guaranteeing the error of RMFA. We conducted toy experiments to demonstrate the efficiency of RMFA and ppSBN, and experiments on long range arena (LRA) benchmark to validate the acceleration and accuracy of Macformer with different dot-product kernels. Experiment results of Macformer are consistent with our theoretical analysis.

This topical review article reports rapid progress on the generalization and application of entanglement in non-Hermitian free-fermion quantum systems. We begin by examining the realization of non-Hermitian quantum systems through the Lindblad master equation, alongside a review of typical non-Hermitian free-fermion systems that exhibit unique features. A pedagogical discussion is provided on the relationship between entanglement quantities and the correlation matrix in Hermitian systems. Building on this foundation, we focus on how entanglement concepts are extended to non-Hermitian systems from their Hermitian free-fermion counterparts, with a review of the general properties that emerge. Finally, we highlight various concrete studies, demonstrating that entanglement entropy remains a powerful diagnostic tool for characterizing non-Hermitian physics. The entanglement spectrum also reflects the topological characteristics of non-Hermitian topological systems, while unique non-Hermitian entanglement behaviors are also discussed. The review is concluded with several future directions. Through this review, we hope to provide a useful guide for researchers who are interested in entanglement in non-Hermitian quantum systems., Comment: version 2; a short review. ~14p, 1figure

In high-energy physics, anti-neutrons ($\bar{n}$) are fundamental particles that frequently appear as final-state particles, and the reconstruction of their kinematic properties provides an important probe for understanding the governing principles. However, this confronts significant challenges instrumentally with the electromagnetic calorimeter (EMC), a typical experimental sensor but recovering the information of incident $\bar{n}$ insufficiently. In this study, we introduce Vision Calorimeter (ViC), a baseline method for anti-neutron reconstruction that leverages deep learning detectors to analyze the implicit relationships between EMC responses and incident $\bar{n}$ characteristics. Our motivation lies in that energy distributions of $\bar{n}$ samples deposited in the EMC cell arrays embody rich contextual information. Converted to 2-D images, such contextual energy distributions can be used to predict the status of $\bar{n}$ ($i.e.$, incident position and momentum) through a deep learning detector along with pseudo bounding boxes and a specified training objective. Experimental results demonstrate that ViC substantially outperforms the conventional reconstruction approach, reducing the prediction error of incident position by 42.81% (from 17.31$^{\circ}$ to 9.90$^{\circ}$). More importantly, this study for the first time realizes the measurement of incident $\bar{n}$ momentum, underscoring the potential of deep learning detectors for particle reconstruction. Code is available at https://github.com/yuhongtian17/ViC.

The labelling difficulty has been a longstanding problem in deep image matting. To escape from fine labels, this work explores using rough annotations such as trimaps coarsely indicating the foreground/background as supervision. We present that the cooperation between learned semantics from indicated known regions and proper assumed matting rules can help infer alpha values at transition areas. Inspired by the nonlocal principle in traditional image matting, we build a directional distance consistency loss (DDC loss) at each pixel neighborhood to constrain the alpha values conditioned on the input image. DDC loss forces the distance of similar pairs on the alpha matte and on its corresponding image to be consistent. In this way, the alpha values can be propagated from learned known regions to unknown transition areas. With only images and trimaps, a matting model can be trained under the supervision of a known loss and the proposed DDC loss. Experiments on AM-2K and P3M-10K dataset show that our paradigm achieves comparable performance with the fine-label-supervised baseline, while sometimes offers even more satisfying results than human-labelled ground truth. Code is available at \url{https://github.com/poppuppy/alpha-free-matting}., Comment: 12 pages, 12 figures

Federated Learning (FL) has emerged as an excellent solution for performing deep learning on different data owners without exchanging raw data. However, statistical heterogeneity in FL presents a key challenge, leading to a phenomenon of skewness in local model parameter distributions that researchers have largely overlooked. In this work, we propose the concept of parameter skew to describe the phenomenon that can substantially affect the accuracy of global model parameter estimation. Additionally, we introduce FedSA, an aggregation strategy to obtain a high-quality global model, to address the implication from parameter skew. Specifically, we categorize parameters into high-dispersion and low-dispersion groups based on the coefficient of variation. For high-dispersion parameters, Micro-Classes (MIC) and Macro-Classes (MAC) represent the dispersion at the micro and macro levels, respectively, forming the foundation of FedSA. To evaluate the effectiveness of FedSA, we conduct extensive experiments with different FL algorithms on three computer vision datasets. FedSA outperforms eight state-of-the-art baselines by about 4.7% in test accuracy.

Open-vocabulary 3D scene understanding (OV-3D) aims to localize and classify novel objects beyond the closed object classes. However, existing approaches and benchmarks primarily focus on the open vocabulary problem within the context of object classes, which is insufficient to provide a holistic evaluation to what extent a model understands the 3D scene. In this paper, we introduce a more challenging task called Generalized Open-Vocabulary 3D Scene Understanding (GOV-3D) to explore the open vocabulary problem beyond object classes. It encompasses an open and diverse set of generalized knowledge, expressed as linguistic queries of fine-grained and object-specific attributes. To this end, we contribute a new benchmark named OpenScan, which consists of 3D object attributes across eight representative linguistic aspects, including affordance, property, material, and more. We further evaluate state-of-the-art OV-3D methods on our OpenScan benchmark, and discover that these methods struggle to comprehend the abstract vocabularies of the GOV-3D task, a challenge that cannot be addressed by simply scaling up object classes during training. We highlight the limitations of existing methodologies and explore a promising direction to overcome the identified shortcomings. Data and code are available at https://github.com/YoujunZhao/OpenScan

Embodied AI is transforming how AI systems interact with the physical world, yet existing datasets are inadequate for developing versatile, general-purpose agents. These limitations include a lack of standardized formats, insufficient data diversity, and inadequate data volume. To address these issues, we introduce ARIO (All Robots In One), a new data standard that enhances existing datasets by offering a unified data format, comprehensive sensory modalities, and a combination of real-world and simulated data. ARIO aims to improve the training of embodied AI agents, increasing their robustness and adaptability across various tasks and environments. Building upon the proposed new standard, we present a large-scale unified ARIO dataset, comprising approximately 3 million episodes collected from 258 series and 321,064 tasks. The ARIO standard and dataset represent a significant step towards bridging the gaps of existing data resources. By providing a cohesive framework for data collection and representation, ARIO paves the way for the development of more powerful and versatile embodied AI agents, capable of navigating and interacting with the physical world in increasingly complex and diverse ways. The project is available on https://imaei.github.io/project_pages/ario/, Comment: Project website: https://imaei.github.io/project_pages/ario/

Currently, Audio Language Models (ALMs) are rapidly advancing due to the developments in large language models and audio neural codecs. These ALMs have significantly lowered the barrier to creating deepfake audio, generating highly realistic and diverse types of deepfake audio, which pose severe threats to society. Consequently, effective audio deepfake detection technologies to detect ALM-based audio have become increasingly critical. This paper investigate the effectiveness of current countermeasure (CM) against ALM-based audio. Specifically, we collect 12 types of the latest ALM-based deepfake audio and utilizing the latest CMs to evaluate. Our findings reveal that the latest codec-trained CM can effectively detect ALM-based audio, achieving 0% equal error rate under most ALM test conditions, which exceeded our expectations. This indicates promising directions for future research in ALM-based deepfake audio detection.

Large language models (LLMs) have unlocked a plethora of powerful applications at the network edge, such as intelligent personal assistants. Data privacy and security concerns have prompted a shift towards edge-based fine-tuning of personal LLMs, away from cloud reliance. However, this raises issues of computational intensity and resource scarcity, hindering training efficiency and feasibility. While current studies investigate parameter-efficient fine-tuning (PEFT) techniques to mitigate resource constraints, our analysis indicates that these techniques are not sufficiently resource-efficient for edge devices. To tackle these challenges, we propose Pluto and Charon (PAC), a time and memory efficient collaborative edge AI framework for personal LLMs fine-tuning. PAC breaks the resource wall of personal LLMs fine-tuning with a sophisticated algorithm-system co-design. (1) Algorithmically, PAC implements a personal LLMs fine-tuning technique that is efficient in terms of parameters, time, and memory. It utilizes Parallel Adapters to circumvent the need for a full backward pass through the LLM backbone. Additionally, an activation cache mechanism further streamlining the process by negating the necessity for repeated forward passes across multiple epochs. (2) Systematically, PAC leverages edge devices in close proximity, pooling them as a collective resource for in-situ personal LLMs fine-tuning, utilizing a hybrid data and pipeline parallelism to orchestrate distributed training. The use of the activation cache eliminates the need for forward pass through the LLM backbone,enabling exclusive fine-tuning of the Parallel Adapters using data parallelism. Extensive evaluation based on prototype implementation demonstrates that PAC remarkably outperforms state-of-the-art approaches, achieving up to 8.64x end-to-end speedup and up to 88.16% reduction in memory footprint., Comment: Accepted by The 53rd International Conference on Parallel Processing (ICPP'24)

Parameters in an effective field theory can be subject to certain positivity bounds if one requires a UV completion that obeys the fundamental principles of quantum field theory. These bounds are relatively straight forward at the tree level, but would become more obscure when loop effects are important. Using scalar theories as examples, we carefully exam the positivity bounds in a case where the leading contribution to a forward elastic amplitude arises at the one-loop level, and point out certain subtleties in terms of the implications of positivity bounds on the theory parameter space. In particular, the one-loop generated dimension-8 operator coefficients (that would be positive if generated at the tree level), as well as their $\beta$-functions are generally not subject to positivity bounds as they might correspond to interference terms of the cross sections under the optical theorem, which could have either sign. A strict positivity bound can only be implied when all contributions at the same loop order are considered, including the ones from dim-4 and dim-6 operator coefficients, which have important effects at the one-loop level. Our results may have important implications on the robustness of experimental tests of positivity bounds., Comment: 22 pages, 4 figures