Search

We present TimeWalker, a novel framework that models realistic, full-scale 3D head avatars of a person on lifelong scale. Unlike current human head avatar pipelines that capture identity at the momentary level(e.g., instant photography or short videos), TimeWalker constructs a person's comprehensive identity from unstructured data collection over his/her various life stages, offering a paradigm to achieve full reconstruction and animation of that person at different moments of life. At the heart of TimeWalker's success is a novel neural parametric model that learns personalized representation with the disentanglement of shape, expression, and appearance across ages. Central to our methodology are the concepts of two aspects: (1) We track back to the principle of modeling a person's identity in an additive combination of average head representation in the canonical space, and moment-specific head attribute representations driven from a set of neural head basis. To learn the set of head basis that could represent the comprehensive head variations in a compact manner, we propose a Dynamic Neural Basis-Blending Module (Dynamo). It dynamically adjusts the number and blend weights of neural head bases, according to both shared and specific traits of the target person over ages. (2) Dynamic 2D Gaussian Splatting (DNA-2DGS), an extension of Gaussian splatting representation, to model head motion deformations like facial expressions without losing the realism of rendering and reconstruction. DNA-2DGS includes a set of controllable 2D oriented planar Gaussian disks that utilize the priors from parametric model, and move/rotate with the change of expression. Through extensive experimental evaluations, we show TimeWalker's ability to reconstruct and animate avatars across decoupled dimensions with realistic rendering effects, demonstrating a way to achieve personalized 'time traveling' in a breeze., Comment: Project Page: https://timewalker2024.github.io/timewalker.github.io/ , Video: https://www.youtube.com/watch?v=x8cpOVMY_ko

Natural orbital functional (NOF) theory offers a promising approach for studying strongly correlated systems at an affordable computational cost, with an accuracy comparable to highly demanding wavefunction-based methods. However, its widespread adoption in cases involving a large number of correlated electrons has been limited by the extensive iterations required for convergence. In this work, we present a disruptive approach that embeds the techniques used for optimization in deep learning within the NOF calculation, representing a substantial advance in the scale of accessible systems. The revamped procedure is based on the adaptive momentum technique for orbital optimization, alternated with the optimization of the occupation numbers, significantly improving the computational feasibility of challenging calculations. We demonstrate this with three examples that involve a large number of electrons: (i) the symmetric dissociation of a large hydrogen cluster, (ii) an analysis of occupancies distribution in fullerenes, and (iii) a study of the singlet-triplet energy gap in linear acenes. Notably, the first example serves as an ideal model for a strongly correlated Mott insulator, featuring 1000 electrons and illustrating a metal-to-insulator transition, surpassing the largest calculations of strongly correlated electrons reported to date. We anticipate that this work will enable the practical application of NOFs to increasingly complex and intriguing systems, leveraging the method's inherent scalability and accuracy., Comment: 9 pages, 4 figures

Fairness in both machine learning (ML) predictions and human decisions is critical, with ML models prone to algorithmic and data bias, and human decisions affected by subjectivity and cognitive bias. This study investigates fairness using a real-world university admission dataset with 870 profiles, leveraging three ML models, namely XGB, Bi-LSTM, and KNN. Textual features are encoded with BERT embeddings. For individual fairness, we assess decision consistency among experts with varied backgrounds and ML models, using a consistency score. Results show ML models outperform humans in fairness by 14.08% to 18.79%. For group fairness, we propose a gender-debiasing pipeline and demonstrate its efficacy in removing gender-specific language without compromising prediction performance. Post-debiasing, all models maintain or improve their classification accuracy, validating the hypothesis that fairness and performance can coexist. Our findings highlight ML's potential to enhance fairness in admissions while maintaining high accuracy, advocating a hybrid approach combining human judgement and ML models.

Visual generation models have achieved remarkable progress in computer graphics applications but still face significant challenges in real-world deployment. Current assessment approaches for visual generation tasks typically follow an isolated three-phase framework: test input collection, model output generation, and user assessment. These fashions suffer from fixed coverage, evolving difficulty, and data leakage risks, limiting their effectiveness in comprehensively evaluating increasingly complex generation models. To address these limitations, we propose DyEval, an LLM-powered dynamic interactive visual assessment framework that facilitates collaborative evaluation between humans and generative models for text-to-image systems. DyEval features an intuitive visual interface that enables users to interactively explore and analyze model behaviors, while adaptively generating hierarchical, fine-grained, and diverse textual inputs to continuously probe the capability boundaries of the models based on their feedback. Additionally, to provide interpretable analysis for users to further improve tested models, we develop a contextual reflection module that mines failure triggers of test inputs and reflects model potential failure patterns supporting in-depth analysis using the logical reasoning ability of LLM. Qualitative and quantitative experiments demonstrate that DyEval can effectively help users identify max up to 2.56 times generation failures than conventional methods, and uncover complex and rare failure patterns, such as issues with pronoun generation and specific cultural context generation. Our framework provides valuable insights for improving generative models and has broad implications for advancing the reliability and capabilities of visual generation systems across various domains., Comment: Under Review

Diffusion Transformers (DiTs) have exhibited robust capabilities in image generation tasks. However, accurate text-guided image editing for multimodal DiTs (MM-DiTs) still poses a significant challenge. Unlike UNet-based structures that could utilize self/cross-attention maps for semantic editing, MM-DiTs inherently lack support for explicit and consistent incorporated text guidance, resulting in semantic misalignment between the edited results and texts. In this study, we disclose the sensitivity of different attention heads to different image semantics within MM-DiTs and introduce HeadRouter, a training-free image editing framework that edits the source image by adaptively routing the text guidance to different attention heads in MM-DiTs. Furthermore, we present a dual-token refinement module to refine text/image token representations for precise semantic guidance and accurate region expression. Experimental results on multiple benchmarks demonstrate HeadRouter's performance in terms of editing fidelity and image quality.

Large Language Models are prone to off-topic misuse, where users may prompt these models to perform tasks beyond their intended scope. Current guardrails, which often rely on curated examples or custom classifiers, suffer from high false-positive rates, limited adaptability, and the impracticality of requiring real-world data that is not available in pre-production. In this paper, we introduce a flexible, data-free guardrail development methodology that addresses these challenges. By thoroughly defining the problem space qualitatively and passing this to an LLM to generate diverse prompts, we construct a synthetic dataset to benchmark and train off-topic guardrails that outperform heuristic approaches. Additionally, by framing the task as classifying whether the user prompt is relevant with respect to the system prompt, our guardrails effectively generalize to other misuse categories, including jailbreak and harmful prompts. Lastly, we further contribute to the field by open-sourcing both the synthetic dataset and the off-topic guardrail models, providing valuable resources for developing guardrails in pre-production environments and supporting future research and development in LLM safety., Comment: 8 pages, 5 figures

Generalist vision language models (VLMs) have made significant strides in computer vision, but they fall short in specialized fields like healthcare, where expert knowledge is essential. In traditional computer vision tasks, creative or approximate answers may be acceptable, but in healthcare, precision is paramount.Current large multimodal models like Gemini and GPT-4o are insufficient for medical tasks due to their reliance on memorized internet knowledge rather than the nuanced expertise required in healthcare. VLMs are usually trained in three stages: vision pre-training, vision-language pre-training, and instruction fine-tuning (IFT). IFT has been typically applied using a mixture of generic and healthcare data. In contrast, we propose that for medical VLMs, a fourth stage of specialized IFT is necessary, which focuses on medical data and includes information from domain expert models. Domain expert models developed for medical use are crucial because they are specifically trained for certain clinical tasks, e.g. to detect tumors and classify abnormalities through segmentation and classification, which learn fine-grained features of medical data$-$features that are often too intricate for a VLM to capture effectively especially in radiology. This paper introduces a new framework, VILA-M3, for medical VLMs that utilizes domain knowledge via expert models. Through our experiments, we show an improved state-of-the-art (SOTA) performance with an average improvement of ~9% over the prior SOTA model Med-Gemini and ~6% over models trained on the specific tasks. Our approach emphasizes the importance of domain expertise in creating precise, reliable VLMs for medical applications.

We carried out a project involving the systematic analysis of microlensing data from the Korea Microlensing Telescope Network survey. The aim of this project is to identify lensing events with complex anomaly features that are difficult to explain using standard binary-lens or binary-source models. Our investigation reveals that the light curves of microlensing events KMT-2021-BLG-0284, KMT-2022-BLG-2480, and KMT-2024-BLG-0412 display highly complex patterns with three or more anomaly features. These features cannot be adequately explained by a binary-lens (2L1S) model alone. However, the 2L1S model can effectively describe certain segments of the light curve. By incorporating an additional source into the modeling, we identified a comprehensive model that accounts for all the observed anomaly features. Bayesian analysis, based on constraints provided by lensing observables, indicates that the lenses of KMT-2021-BLG-0284 and KMT-2024-BLG-0412 are binary systems composed of M dwarfs. For KMT-2022-BLG-2480, the primary lens is an early K-type main-sequence star with an M dwarf companion. The lenses of KMT-2021-BLG-0284 and KMT-2024-BLG-0412 are likely located in the bulge, whereas the lens of KMT-2022-BLG-2480 is more likely situated in the disk. In all events, the binary stars of the sources have similar magnitudes due to a detection bias favoring binary source events with a relatively bright secondary source star, which increases detection efficiency., Comment: 9 pages, 9 figures

Current implementations of Bayesian Additive Regression Trees (BART) are based on axis-aligned decision rules that recursively partition the feature space using a single feature at a time. Several authors have demonstrated that oblique trees, whose decision rules are based on linear combinations of features, can sometimes yield better predictions than axis-aligned trees and exhibit excellent theoretical properties. We develop an oblique version of BART that leverages a data-adaptive decision rule prior that recursively partitions the feature space along random hyperplanes. Using several synthetic and real-world benchmark datasets, we systematically compared our oblique BART implementation to axis-aligned BART and other tree ensemble methods, finding that oblique BART was competitive with -- and sometimes much better than -- those methods.

We study the hydrodynamic theories with approximate symmetries in the recently developed effective action approach on the Schwinger-Keldysh (SK) contour. We employ the method of spurious symmetry transformation for small explicit symmetry-breaking parameters to systematically constrain symmetry-breaking effects in the non-equilibrium effective action for hydrodynamics. We apply our method to the hydrodynamic theory of chiral symmetry in Quantum Chromodynamics (QCD) at finite temperature and density and its explicit breaking by quark masses. We show that the spurious symmetry and the Kubo-Martin-Schwinger (KMS) relation dictate that the Ward-Takahashi identity for the axial symmetry, i.e., the partial conservation of axial vector current (PCAC) relation, contains a relaxational term proportional to the axial chemical potential, whose kinetic coefficient is at least of the second order in the quark mass. In the phase where the chiral symmetry is spontaneously broken, and the pseudo-Nambu-Goldstone pions appear as hydrodynamic variables, this relaxation effect is subleading compared to the conventional pion mass term in the PCAC relation, which is of the first order in the quark mass. On the other hand, in the chiral symmetry-restored phase, we show that our relaxation term, which is of the second order in the quark mass, becomes the leading contribution to the axial charge relaxation. Therefore, the leading axial charge relaxation mechanism is parametrically different in the quark mass across a chiral phase transition., Comment: 34 pages, 0 figures

Although it is an extremely effective, easy-to-use, and increasingly popular tool for nonparametric regression, the Bayesian Additive Regression Trees (BART) model is limited by the fact that it can only produce discontinuous output. Initial attempts to overcome this limitation were based on regression trees that output Gaussian Processes instead of constants. Unfortunately, implementations of these extensions cannot scale to large datasets. We propose ridgeBART, an extension of BART built with trees that output linear combinations of ridge functions (i.e., a composition of an affine transformation of the inputs and non-linearity); that is, we build a Bayesian ensemble of localized neural networks with a single hidden layer. We develop a new MCMC sampler that updates trees in linear time and establish nearly minimax-optimal posterior contraction rates for estimating Sobolev and piecewise anisotropic Sobolev functions. We demonstrate ridgeBART's effectiveness on synthetic data and use it to estimate the probability that a professional basketball player makes a shot from any location on the court in a spatially smooth fashion.

Supporting the health and well-being of dynamic populations around the world requires governmental agencies, organizations and researchers to understand and reason over complex relationships between human behavior and local contexts in order to identify high-risk groups and strategically allocate limited resources. Traditional approaches to these classes of problems often entail developing manually curated, task-specific features and models to represent human behavior and the natural and built environment, which can be challenging to adapt to new, or even, related tasks. To address this, we introduce a Population Dynamics Foundation Model (PDFM) that aims to capture the relationships between diverse data modalities and is applicable to a broad range of geospatial tasks. We first construct a geo-indexed dataset for postal codes and counties across the United States, capturing rich aggregated information on human behavior from maps, busyness, and aggregated search trends, and environmental factors such as weather and air quality. We then model this data and the complex relationships between locations using a graph neural network, producing embeddings that can be adapted to a wide range of downstream tasks using relatively simple models. We evaluate the effectiveness of our approach by benchmarking it on 27 downstream tasks spanning three distinct domains: health indicators, socioeconomic factors, and environmental measurements. The approach achieves state-of-the-art performance on all 27 geospatial interpolation tasks, and on 25 out of the 27 extrapolation and super-resolution tasks. We combined the PDFM with a state-of-the-art forecasting foundation model, TimesFM, to predict unemployment and poverty, achieving performance that surpasses fully supervised forecasting. The full set of embeddings and sample code are publicly available for researchers., Comment: 28 pages, 16 figures, preprint; v2: updated github url

We present an empirical study investigating how specific properties of preference datasets, such as mixed-quality or noisy data, affect the performance of Preference Optimization (PO) algorithms. Our experiments, conducted in MuJoCo environments, reveal several scenarios where state-of-the-art PO methods experience significant drops in performance. To address this issue, we introduce a novel PO framework based on mirror descent, which can recover existing methods like Direct Preference Optimization (DPO) and Odds-Ratio Preference Optimization (ORPO) for specific choices of the mirror map. Within this framework, we employ evolutionary strategies to discover new loss functions capable of handling the identified problematic scenarios. These new loss functions lead to significant performance improvements over DPO and ORPO across several tasks. Additionally, we demonstrate the generalization capability of our approach by applying the discovered loss functions to fine-tuning large language models using mixed-quality data, where they outperform ORPO.

In the digital age, users store personal data in corporate databases, making data security central to enterprise management. Given the extensive attack surface, assets face challenges like weak authentication, vulnerabilities, and malware. Attackers may exploit vulnerabilities to gain unauthorized access, masquerading as legitimate users. Such attacks can lead to privacy breaches, business disruption, financial losses, and reputational damage. Complex attack vectors blur lines between insider and external threats. To address this, we introduce the IDU-Detector, integrating Intrusion Detection Systems (IDS) with User and Entity Behavior Analytics (UEBA). This integration monitors unauthorized access, bridges system gaps, ensures continuous monitoring, and enhances threat identification. Existing insider threat datasets lack depth and coverage of diverse attack vectors. This hinders detection technologies from addressing complex attack surfaces. We propose new, diverse datasets covering more attack scenarios, enhancing detection technologies. Testing our framework, the IDU-Detector achieved average accuracies of 98.96% and 99.12%. These results show effectiveness in detecting attacks, improving security and response speed, and providing higher asset safety assurance.

We analysed microlensing data to uncover the nature of the anomaly that appeared near the peak of the short-timescale microlensing event KMT-2024-BLG-1044. Despite the anomaly's brief duration of less than a day, it was densely observed through high-cadence monitoring conducted by the KMTNet survey. Detailed modelling of the light curve confirmed the planetary origin of the anomaly and revealed two possible solutions, due to an inner--outer degeneracy. The two solutions provide different measured planet parameters: $(s, q)_{\rm inner} = [1.0883 \pm 0.0027, (3.125 \pm 0.248)\times 10^{-4}]$ for the inner solutions and $(s, q)_{\rm outer} = [1.0327 \pm 0.0054, (3.350 \pm 0.316)\times 10^{-4}]$ for the outer solutions. Using Bayesian analysis with constraints provided by the short event timescale ($t_{\rm E} \sim 9.1$~day) and the small angular Einstein radius ($\theta_{\rm E}\sim 0.16$~mas for the inner solution and $\sim 0.10$~mas for the outer solutio), we determined that the lens is a planetary system consisting of a host near the boundary between a star and a brown dwarf and a planet with a mass lower than that of Uranus. The discovery of the planetary system highlights the crucial role of the microlensing technique in detecting planets that orbit substellar brown dwarfs or very low-mass stars., Comment: 8 pages, 10 figures

In today's high-pressure and isolated society, the demand for emotional support has surged, necessitating innovative solutions. Socially Assistive Robots (SARs) offer a technological approach to providing emotional assistance by leveraging advanced robotics, artificial intelligence, and sensor technologies. This study explores the development of an emotional support robot designed to detect and respond to human emotions, particularly sadness, through facial recognition and gesture analysis. Utilising the Lego Mindstorms Robotic Kit, Raspberry Pi 4, and various Python libraries, the robot is capable of delivering empathetic interactions, including comforting hugs and AI-generated conversations. Experimental findings highlight the robot's effective facial recognition accuracy, user interaction, and hug feedback mechanisms. These results demonstrate the feasibility of using SARs for emotional support, showcasing their potential features and functions. This research underscores the promise of SARs in providing innovative emotional assistance and enhancing human-robot interaction., Comment: Part of proceedings of 6th International Conference AsiaHaptics 2024

TOI-2818 b is a hot Jupiter orbiting a slightly evolved G-type star on a 4.04-day orbit that shows transit timing variations (TTVs) suggestive of a decreasing orbital period. In the most recent year of TESS observations, transits were observed $\sim$8 minutes earlier than expected for a constant period. The implied orbital decay rate is $1.35 \pm 0.25$ s yr$^{-1}$, too fast to be explained by tidal dissipation even considering the evolved nature of the host star. Radial velocity monitoring rules out the possibility that the apparent change in period is due to a steady acceleration of the star by a long-period companion. Apsidal precession due to the tidal distortion of the planet is also physically implausible. The most plausible explanation for the TTVs appears to be gravitational perturbations from a hitherto undetected planet with mass $\lesssim$$10\,M_\oplus$ that is in (or near) a mean-motion resonance with the hot Jupiter. Such a planet could be responsible for the observed TTVs while avoiding detection with the available radial velocity and transit data., Comment: 10 pages, 2 figures, 3 tables, submitted to ApJL

Neural networks are traditionally trained under the assumption that data come from a stationary distribution. However, settings which violate this assumption are becoming more popular; examples include supervised learning under distributional shifts, reinforcement learning, continual learning and non-stationary contextual bandits. In this work we introduce a novel learning approach that automatically models and adapts to non-stationarity, via an Ornstein-Uhlenbeck process with an adaptive drift parameter. The adaptive drift tends to draw the parameters towards the initialisation distribution, so the approach can be understood as a form of soft parameter reset. We show empirically that our approach performs well in non-stationary supervised and off-policy reinforcement learning settings.

Band alignment of metal contacts to 2D semiconductors often deviate from the ideal Shottky-Mott (SM) rule due to the non-ideal factors such as the formation of interface dipole and metal-induced gap states (MIGS). Although MIGS can be strongly suppressed using van der Waals (vdW) contact engineering, the interface dipole is hard to eliminate due to the electronegativity difference of the two contacting materials. Here we show that interface dipole can be practically eliminated in 2D semiconducting MoSi$_2$N$_4$ when contacted by its homologous metallic counterpart MoSi$_2$N$_4$(MoN)$_n$ ($n = 1-4$). The SiN outer sublayers, simultaneously present in both MoSi$_2$N$_4$ and MoSi$_2$N$_4$(MoN)$_n$, creates nearly equal charge `push-back' effect at the contact interface. This nearly symmetrical charge redistribution leads to zero net electron transfer across the interface, resulting in a \emph{zero-dipole} contact. Intriguingly, we show that even in the extreme close-contact case where MoSi$_2$N$_4$(MoN) is arbitrarily pushed towards MoSi$_2$N$_4$ with extremely small interlayer distance, the interface dipole remains practically zero. Such \emph{zero-dipole} Schottky contact represents a peculiar case where the SM rule, usually expected to occur only in the non-interacting regime, manifests in MoSi$_2$N$_4$/MoSi$_2$N$_4$(MoN)$_n$ vdWH even though the constituent monolayers interact strongly. A model for pressure sensing is then proposed based on changing the interlayer distance in MoSi$_2$N$_4$/MoSi$_2$N$_4$(MoN) vdWH.

This document is comprised of a collection of updated preliminary parameters for the key parts of the muon collider. The updated preliminary parameters follow on from the October 2023 Tentative Parameters Report. Particular attention has been given to regions of the facility that are believed to hold greater technical uncertainty in their design and that have a strong impact on the cost and power consumption of the facility. The data is collected from a collaborative spreadsheet and transferred to overleaf.

Personalized medicine (PM) promises to transform healthcare by providing treatments tailored to individual genetic, environmental, and lifestyle factors. However, its high costs and infrastructure demands raise concerns about exacerbating health disparities, especially between high-income countries (HICs) and low- and middle-income countries (LMICs). While HICs benefit from advanced PM applications through AI and genomics, LMICs often lack the resources necessary to adopt these innovations, leading to a widening healthcare divide. This paper explores the financial and ethical challenges of PM implementation, with a focus on ensuring equitable access. It proposes strategies for global collaboration, infrastructure development, and ethical frameworks to support LMICs in adopting PM, aiming to prevent further disparities in healthcare accessibility and outcomes., Comment: 30 pages, 1 figure

Aggregated relative search frequencies offer a unique composite signal reflecting people's habits, concerns, interests, intents, and general information needs, which are not found in other readily available datasets. Temporal search trends have been successfully used in time series modeling across a variety of domains such as infectious diseases, unemployment rates, and retail sales. However, most existing applications require curating specialized datasets of individual keywords, queries, or query clusters, and the search data need to be temporally aligned with the outcome variable of interest. We propose a novel approach for generating an aggregated and anonymized representation of search interest as foundation features at the community level for geospatial modeling. We benchmark these features using spatial datasets across multiple domains. In zip codes with a population greater than 3000 that cover over 95% of the contiguous US population, our models for predicting missing values in a 20% set of holdout counties achieve an average $R^2$ score of 0.74 across 21 health variables, and 0.80 across 6 demographic and environmental variables. Our results demonstrate that these search features can be used for spatial predictions without strict temporal alignment, and that the resulting models outperform spatial interpolation and state of the art methods using satellite imagery features., Comment: 8 pages, 8 figures, presented at the DMLR workshop at ICML 2024

There is a tremendous amount of interest in AI/ML technologies due to the proliferation of generative AI applications such as ChatGPT. This trend has significantly increased demand on GPUs, which are the workhorses for training AI models. Due to the high costs of GPUs and lacking supply, it has become of interest to optimize GPU usage in HPC centers. MIT Lincoln Laboratory Supercomputing Center (LLSC) has developed an easy-to-use GPU sharing feature supported by LLSC-developed tools including LLsub and LLMapReduce. This approach overcomes some of the limitations with the existing methods for GPU sharing. This allows users to apply GPU sharing whenever possible while they are developing their AI/ML models and/or doing parametric study on their AI models or executing other GPU applications. Based on our initial experimental results with GPU sharing, GPU sharing with triples mode is easy to use and achieved significant improvement in GPU usage and throughput performance for certain types of AI applications.

Supervised fine-tuning (SFT) and alignment of large language models (LLMs) are key steps in providing a good user experience. However, the concept of an appropriate alignment is inherently application-dependent, and current methods often rely on heuristic choices to drive the optimization. In this work, we formulate SFT and alignment as a constrained optimization problem, where the LLM is trained on a task while being required to meet application-specific requirements, without resorting to heuristics. To solve this, we propose Lagrange Large Language Models (L3Ms), which employ logarithmic barriers to enforce the constraints. This approach allows for the customization of L3Ms across diverse applications while avoiding heuristic-driven processes. We demonstrate experimentally the versatility and efficacy of L3Ms in achieving tailored alignments for various applications.

The increasing use and cost of high performance computing (HPC) requires new easy-to-use tools to enable HPC users and HPC systems engineers to transparently understand the utilization of resources. The MIT Lincoln Laboratory Supercomputing Center (LLSC) has developed a simple command, LLload, to monitor and characterize HPC workloads. LLload plays an important role in identifying opportunities for better utilization of compute resources. LLload can be used to monitor jobs both programmatically and interactively. LLload can characterize users' jobs using various LLload options to achieve better efficiency. This information can be used to inform the user to optimize HPC workloads and improve both CPU and GPU utilization. This includes improvements using judicious oversubscription of the computing resources. Preliminary results suggest significant improvement in GPU utilization and overall throughput performance with GPU overloading in some cases. By enabling users to observe and fix incorrect job submission and/or inappropriate execution setups, LLload can increase the resource usage and improve the overall throughput performance. LLload is a light-weight, easy-to-use tool for both HPC users and HPC systems engineers to monitor HPC workloads to improve system utilization and efficiency.

In this paper, we present \textbf{\textit{FasterCache}}, a novel training-free strategy designed to accelerate the inference of video diffusion models with high-quality generation. By analyzing existing cache-based methods, we observe that \textit{directly reusing adjacent-step features degrades video quality due to the loss of subtle variations}. We further perform a pioneering investigation of the acceleration potential of classifier-free guidance (CFG) and reveal significant redundancy between conditional and unconditional features within the same timestep. Capitalizing on these observations, we introduce FasterCache to substantially accelerate diffusion-based video generation. Our key contributions include a dynamic feature reuse strategy that preserves both feature distinction and temporal continuity, and CFG-Cache which optimizes the reuse of conditional and unconditional outputs to further enhance inference speed without compromising video quality. We empirically evaluate FasterCache on recent video diffusion models. Experimental results show that FasterCache can significantly accelerate video generation (\eg 1.67$\times$ speedup on Vchitect-2.0) while keeping video quality comparable to the baseline, and consistently outperform existing methods in both inference speed and video quality.

In this paper, we propose a model-agnostic cost-effective approach to developing bilingual base large language models (LLMs) to support English and any target language. The method includes vocabulary expansion, initialization of new embeddings, model training and evaluation. We performed our experiments with three languages, each using a non-Latin script - Ukrainian, Arabic, and Georgian. Our approach demonstrates improved language performance while reducing computational costs. It mitigates the disproportionate penalization of underrepresented languages, promoting fairness and minimizing adverse phenomena such as code-switching and broken grammar. Additionally, we introduce new metrics to evaluate language quality, revealing that vocabulary size significantly impacts the quality of generated text.

This paper presents a systematic review of the infrastructure requirements for deploying Large Language Models (LLMs) on-device within the context of small and medium-sized enterprises (SMEs), focusing on both hardware and software perspectives. From the hardware viewpoint, we discuss the utilization of processing units like GPUs and TPUs, efficient memory and storage solutions, and strategies for effective deployment, addressing the challenges of limited computational resources typical in SME settings. From the software perspective, we explore framework compatibility, operating system optimization, and the use of specialized libraries tailored for resource-constrained environments. The review is structured to first identify the unique challenges faced by SMEs in deploying LLMs on-device, followed by an exploration of the opportunities that both hardware innovations and software adaptations offer to overcome these obstacles. Such a structured review provides practical insights, contributing significantly to the community by enhancing the technological resilience of SMEs in integrating LLMs., Comment: 9 pages, 1 figure. The work is supported by the SIT-NVIDIA Joint AI Centre

We introduce BiGR, a novel conditional image generation model using compact binary latent codes for generative training, focusing on enhancing both generation and representation capabilities. BiGR is the first conditional generative model that unifies generation and discrimination within the same framework. BiGR features a binary tokenizer, a masked modeling mechanism, and a binary transcoder for binary code prediction. Additionally, we introduce a novel entropy-ordered sampling method to enable efficient image generation. Extensive experiments validate BiGR's superior performance in generation quality, as measured by FID-50k, and representation capabilities, as evidenced by linear-probe accuracy. Moreover, BiGR showcases zero-shot generalization across various vision tasks, enabling applications such as image inpainting, outpainting, editing, interpolation, and enrichment, without the need for structural modifications. Our findings suggest that BiGR unifies generative and discriminative tasks effectively, paving the way for further advancements in the field., Comment: Project page: https://haoosz.github.io/BiGR

We present the discovery of TOI-7041 b (TIC 201175570 b), a hot Saturn transiting a red giant star with measurable stellar oscillations. We observe solar-like oscillations in TOI-7041 with a frequency of maximum power of $\nu_{\rm max} = 218.50\pm2.23$ $\mu$Hz and a large frequency separation of $\Delta\nu = 16.5282\pm0.0186$ $\mu$Hz. Our asteroseismic analysis indicates that TOI-7041 has a radius of $4.10 \pm 0.06$(stat) $\pm$ 0.05(sys) $R_\odot$, making it one of the largest stars around which a transiting planet has been discovered with the Transiting Exoplanet Survey Satellite (TESS), and the mission's first oscillating red giant with a transiting planet. TOI-7041 b has an orbital period of $9.691 \pm 0.006$ days and a low eccentricity of $e = 0.04 \pm 0.04$. We measure a planet radius of $1.02 \pm 0.03$ $R_J$ with photometry from TESS, and a planet mass of $0.36 \pm 0.16$ $M_J$ ($114 \pm 51$ $M_\oplus$) with ground-based radial velocity measurements. TOI-7041 b appears less inflated than similar systems receiving equivalent incident flux, and its circular orbit indicates that it is not undergoing tidal heating due to circularization. The asteroseismic analysis of the host star provides some of the tightest constraints on stellar properties for a TESS planet host and enables precise characterization of the hot Saturn. This system joins a small number of TESS-discovered exoplanets orbiting stars that exhibit clear stellar oscillations and indicates that extended TESS observations of evolved stars will similarly provide a path to improved exoplanet characterization., Comment: 16 pages, 9 figures, 3 tables

This paper is a preliminary study on the security and design of cryptosystems using Gabidulin-Kronecker Product Codes. In particular, we point out the design impracticality of the system, and propose ways to improve it.

Recent advancements in diffusion models have led to significant improvements in the generation and animation of 4D full-body human-object interactions (HOI). Nevertheless, existing methods primarily focus on SMPL-based motion generation, which is limited by the scarcity of realistic large-scale interaction data. This constraint affects their ability to create everyday HOI scenes. This paper addresses this challenge using a zero-shot approach with a pre-trained diffusion model. Despite this potential, achieving our goals is difficult due to the diffusion model's lack of understanding of ''where'' and ''how'' objects interact with the human body. To tackle these issues, we introduce AvatarGO, a novel framework designed to generate animatable 4D HOI scenes directly from textual inputs. Specifically, 1) for the ''where'' challenge, we propose LLM-guided contact retargeting, which employs Lang-SAM to identify the contact body part from text prompts, ensuring precise representation of human-object spatial relations. 2) For the ''how'' challenge, we introduce correspondence-aware motion optimization that constructs motion fields for both human and object models using the linear blend skinning function from SMPL-X. Our framework not only generates coherent compositional motions, but also exhibits greater robustness in handling penetration issues. Extensive experiments with existing methods validate AvatarGO's superior generation and animation capabilities on a variety of human-object pairs and diverse poses. As the first attempt to synthesize 4D avatars with object interactions, we hope AvatarGO could open new doors for human-centric 4D content creation., Comment: Project page: https://yukangcao.github.io/AvatarGO/

We propose SymDiff, a novel method for constructing equivariant diffusion models using the recently introduced framework of stochastic symmetrisation. SymDiff resembles a learned data augmentation that is deployed at sampling time, and is lightweight, computationally efficient, and easy to implement on top of arbitrary off-the-shelf models. Notably, in contrast to previous work, SymDiff typically does not require any neural network components that are intrinsically equivariant, avoiding the need for complex parameterizations and the use of higher-order geometric features. Instead, our method can leverage highly scalable modern architectures as drop-in replacements for these more constrained alternatives. We show that this additional flexibility yields significant empirical benefit on $\mathrm{E}(3)$-equivariant molecular generation. To the best of our knowledge, this is the first application of symmetrisation to generative modelling, suggesting its potential in this domain more generally.

Predicting time-dependent dynamics of complex systems governed by non-linear partial differential equations (PDEs) with varying parameters and domains is a challenging task motivated by applications across various fields. We introduce a novel family of neural operators based on our Graph Fourier Neural Kernels, designed to learn solution generators for nonlinear PDEs in which the highest-order term is diffusive, across multiple domains and parameters. G-FuNK combines components that are parameter- and domain-adapted with others that are not. The domain-adapted components are constructed using a weighted graph on the discretized domain, where the graph Laplacian approximates the highest-order diffusive term, ensuring boundary condition compliance and capturing the parameter and domain-specific behavior. Meanwhile, the learned components transfer across domains and parameters using our variant Fourier Neural Operators. This approach naturally embeds geometric and directional information, improving generalization to new test domains without need for retraining the network. To handle temporal dynamics, our method incorporates an integrated ODE solver to predict the evolution of the system. Experiments show G-FuNK's capability to accurately approximate heat, reaction diffusion, and cardiac electrophysiology equations across various geometries and anisotropic diffusivity fields. G-FuNK achieves low relative errors on unseen domains and fiber fields, significantly accelerating predictions compared to traditional finite-element solvers.

This paper reviews published research in the field of computer-aided colorization technology. We argue that the colorization task originates from computer graphics, prospers by introducing computer vision, and tends to the fusion of vision and graphics, so we put forward our taxonomy and organize the whole paper chronologically. We extend the existing reconstruction-based colorization evaluation techniques, considering that aesthetic assessment of colored images should be introduced to ensure that colorization satisfies human visual-related requirements and emotions more closely. We perform the colorization aesthetic assessment on seven representative unconditional colorization models and discuss the difference between our assessment and the existing reconstruction-based metrics. Finally, this paper identifies unresolved issues and proposes fruitful areas for future research and development. Access to the project associated with this survey can be obtained at https://github.com/DanielCho-HK/Colorization.

Synthetic tabular data generation has gained significant attention for its potential in data augmentation, software testing and privacy-preserving data sharing. However, most research has primarily focused on larger datasets and evaluating their quality in terms of metrics like column-wise statistical distributions and inter-feature correlations, while often overlooking its utility for data augmentation, particularly for datasets whose data is scarce. In this paper, we propose Tabular Auto-Encoder Generative Adversarial Network (TAEGAN), an improved GAN-based framework for generating high-quality tabular data. Although large language models (LLMs)-based methods represent the state-of-the-art in synthetic tabular data generation, they are often overkill for small datasets due to their extensive size and complexity. TAEGAN employs a masked auto-encoder as the generator, which for the first time introduces the power of self-supervised pre-training in tabular data generation so that essentially exposes the networks to more information. We extensively evaluate TAEGAN against five state-of-the-art synthetic tabular data generation algorithms. Results from 10 datasets show that TAEGAN outperforms existing deep-learning-based tabular data generation models on 9 out of 10 datasets on the machine learning efficacy and achieves superior data augmentation performance on 7 out of 8 smaller datasets.

Real-time supercomputing performance analysis is a critical aspect of evaluating and optimizing computational systems in a dynamic user environment. The operation of supercomputers produce vast quantities of analytic data from multiple sources and of varying types so compiling this data in an efficient matter is critical to the process. MIT Lincoln Laboratory Supercomputing Center has been utilizing the Unity 3D game engine to create a Digital Twin of our supercomputing systems for several years to perform system monitoring. Unity offers robust visualization capabilities making it ideal for creating a sophisticated representation of the computational processes. As we scale the systems to include a diversity of resources such as accelerators and the addition of more users, we need to implement new analysis tools for the monitoring system. The workloads in research continuously change, as does the capability of Unity, and this allows us to adapt our monitoring tools to scale and incorporate features enabling efficient replay of system wide events, user isolation, and machine level granularity. Our system fully takes advantage of the modern capabilities of the Unity Engine in a way that intuitively represents the real time workload performed on a supercomputer. It allows HPC system engineers to quickly diagnose usage related errors with its responsive user interface which scales efficiently with large data sets.

Future electronics require aggressive scaling of channel material thickness while maintaining device performance. Two-dimensional (2D) semiconductors are promising candidates, but despite over two decades of research, experimental performance still lags theoretical expectations. Here, we develop an oxygen-free approach to push the electrical transport of 2D field-effect transistors toward the theoretical phonon-limited intrinsic mobility. We achieve record carrier mobilities of 91 (132) cm2V-1s-1 for mono- (bi-) layer MoS2 transistors on SiO2 substrate. Statistics from over 60 devices confirm that oxygen-free fabrication enhances key figures of merit by more than an order of magnitude. While previous studies suggest that 2D transition metal dichalcogenides such as MoS2 and WS2 are stable in air, we show that short-term ambient exposure can degrade their device performance through irreversible oxygen chemisorption. This study emphasizes the criticality of avoiding oxygen exposure, offering guidance for device manufacturing for fundamental research and practical applications of 2D materials.