Search

Solar energetic particles (SEPs) associated with solar flares and coronal mass ejections (CMEs) are key agents of space weather phenomena, posing severe threats to spacecraft and astronauts. Recent observations by Parker Solar Probe (PSP) indicate that the magnetic flux ropes of a CME can trap energetic particles and act as barriers, preventing other particles from crossing. In this paper, we introduce the novel COCONUT+PARADISE model to investigate the confinement of energetic particles within a flux rope and the effects of cross-field diffusion (CFD) on particle transport in the solar corona, particularly in the presence of a CME. Using the global magnetohydrodynamic coronal model COCONUT, we generate background configurations containing a CME modeled as a Titov-D\'emoulin flux rope (TDFR). We then utilize the particle transport code PARADISE to inject monoenergetic 100 keV protons inside one of the TDFR legs near its footpoint and evolve the particles through the COCONUT backgrounds. To study CFD, we employ two different approaches regarding the perpendicular proton mean free path (MFP): a constant MFP and a Larmor radius-dependent MFP. We contrast these results with those obtained without CFD. While particles remain fully trapped within the TDFR without CFD, we find that even relatively small perpendicular MFP values allow particles on the outer layers to escape. In contrast, the initially interior trapped particles stay largely confined. Finally, we highlight how our model and this paper's results are relevant for future research on particle acceleration and transport in an extended domain encompassing both the corona and inner heliosphere., Comment: 16 pages, 8 figures, accepted for publication by ApJ Letters

The Automated Verification of Textual Claims (AVeriTeC) shared task asks participants to retrieve evidence and predict veracity for real-world claims checked by fact-checkers. Evidence can be found either via a search engine, or via a knowledge store provided by the organisers. Submissions are evaluated using AVeriTeC score, which considers a claim to be accurately verified if and only if both the verdict is correct and retrieved evidence is considered to meet a certain quality threshold. The shared task received 21 submissions, 18 of which surpassed our baseline. The winning team was TUDA_MAI with an AVeriTeC score of 63%. In this paper we describe the shared task, present the full results, and highlight key takeaways from the shared task.

Metabolite biosynthesis is regulated via metabolic pathways, which can be activated and deactivated within organisms. Understanding and identifying an organism's metabolic pathway network is a crucial aspect for various research fields, including crop and life stock breeding, pharmacology, and medicine. The problem of identifying whether a pathway is part of a studied metabolic system is commonly framed as a hyperlink prediction problem. The most important challenge in prediction of metabolic pathways is the sparsity of the labeled data. This challenge can partially be mitigated using metabolite correlation networks which are affected by all active pathways including those that were not confirmed yet in laboratory experiments. Unfortunately, extracting properties that can confirm or refute existence of a metabolic pathway in a particular organism is not a trivial task. In this research, we introduce the Network Auralization Hyperlink Prediction (NetAurHPD) which is a framework that relies on (1) graph auralization to extract and aggregate representations of nodes in metabolite correlation networks and (2) data augmentation method that generates metabolite correlation networks given a subset of chemical reactions defined as hyperlinks. Experiments with metabolites correlation-based networks of tomato pericarp demonstrate promising results for NetAurHPD, compared to alternative methods. Furthermore, the application of data augmentation improved NetAurHPD's learning capabilities and overall performance. Additionally, NetAurHPD outperformed state-of-the-art method in experiments under challenging conditions, and has the potential to be a valuable tool for exploring organisms with limited existing knowledge., Comment: 10 pages, 9 figures, 1 tables, "Submitted to KDD 2025"

LLMs can be used in a variety of code related tasks such as translating from one programming language to another, implementing natural language requirements and code summarization. Artifacts generated by state of the art LLM technology are expected to be useful in the sense that a user will be able to use the LLM generated artifact after a small number of easy modifications. Quantifying this vague notion is challenging and it is thus hard to determine the quality of code related LLM solutions. We refer to evaluation of LLM solutions using LLM judgment as "LLM as a Judge", or LaaJ for short. In this work we introduce a methodology to generate and evaluate LaaJ implementations, utilizing an automatically generated benchmark. The purpose of the benchmark is two fold, namely, it is used both to develop and validate the LaaJs and to validate and test the LLM code related solution using the LaaJs. To that end, we developed an automated benchmark generation engine, which generates code in multiple programming languages for multiple code related tasks and which serves as the input for LaaJ evaluation. We utilize a graph representation, G, of the potential code related generations. The graph vertices are generated artifacts and edges represent possible generations, e.g., the generation of a Java program from its natural language requirements. Utilizing a chain of LLM agents and G we generate code related artifacts. Using cycles in G we formulate expectations on the generated artifacts. Taking advantage of these formulated expectations enables the development and testing of reliable LLM judgement for usefulness of the artifacts generated by the solution. Our approach enables the creation of high quality code task solutions.

The Internet of Things (IoT) is one of the fastest-growing computing industries. By the end of 2027, more than 29 billion devices are expected to be connected. These smart devices can communicate with each other with and without human intervention. This rapid growth has led to the emergence of new types of malware. However, traditional malware detection methods, such as signature-based and heuristic-based techniques, are becoming increasingly ineffective against these new types of malware. Therefore, it has become indispensable to find practical solutions for detecting IoT malware. Machine Learning (ML) and Deep Learning (DL) approaches have proven effective in dealing with these new IoT malware variants, exhibiting high detection rates. In this paper, we bridge the gap in research between the IoT malware analysis and the wide adoption of deep learning in tackling the problems in this domain. As such, we provide a comprehensive review on deep learning based malware analysis across various categories of the IoT domain (i.e. Extended Internet of Things (XIoT)), including Industrial IoT (IIoT), Internet of Medical Things (IoMT), Internet of Vehicles (IoV), and Internet of Battlefield Things (IoBT).

The recent development of fact verification systems with natural logic has enhanced their explainability by aligning claims with evidence through set-theoretic operators, providing faithful justifications. Despite these advancements, such systems often rely on a large amount of training data annotated with natural logic. To address this issue, we propose a zero-shot method that utilizes the generalization capabilities of instruction-tuned large language models. To comprehensively assess the zero-shot capabilities of our method and other fact verification systems, we evaluate all models on both artificial and real-world claims, including multilingual datasets. We also compare our method against other fact verification systems in two setups. First, in the zero-shot generalization setup, we demonstrate that our approach outperforms other systems that were not specifically trained on natural logic data, achieving an average accuracy improvement of 8.96 points over the best-performing baseline. Second, in the zero-shot transfer setup, we show that current systems trained on natural logic data do not generalize well to other domains, and our method outperforms these systems across all datasets with real-world claims., Comment: Accepted to EMNLP 2024

Recent retrieval-augmented models enhance basic methods by building a hierarchical structure over retrieved text chunks through recursive embedding, clustering, and summarization. The most relevant information is then retrieved from both the original text and generated summaries. However, such approaches face limitations with dynamic datasets, where adding or removing documents over time complicates the updating of hierarchical representations formed through clustering. We propose a new algorithm to efficiently maintain the recursive-abstractive tree structure in dynamic datasets, without compromising performance. Additionally, we introduce a novel post-retrieval method that applies query-focused recursive abstractive processing to substantially improve context quality. Our method overcomes the limitations of other approaches by functioning as a black-box post-retrieval layer compatible with any retrieval algorithm. Both algorithms are validated through extensive experiments on real-world datasets, demonstrating their effectiveness in handling dynamic data and improving retrieval performance.

Recent advancements in generative models have revolutionized the field of artificial intelligence, enabling the creation of highly-realistic and detailed images. In this study, we propose a novel Mask Conditional Text-to-Image Generative Model (MCGM) that leverages the power of conditional diffusion models to generate pictures with specific poses. Our model builds upon the success of the Break-a-scene [1] model in generating new scenes using a single image with multiple subjects and incorporates a mask embedding injection that allows the conditioning of the generation process. By introducing this additional level of control, MCGM offers a flexible and intuitive approach for generating specific poses for one or more subjects learned from a single image, empowering users to influence the output based on their requirements. Through extensive experimentation and evaluation, we demonstrate the effectiveness of our proposed model in generating high-quality images that meet predefined mask conditions and improving the current Break-a-scene generative model., Comment: 17 pages, 13 figures, presented at the 5th International Conference on Artificial Intelligence and Machine Learning (CAIML 2024)

Human-like personality traits have recently been discovered in large language models, raising the hypothesis that their (known and as yet undiscovered) biases conform with human latent psychological constructs. While large conversational models may be tricked into answering psychometric questionnaires, the latent psychological constructs of thousands of simpler transformers, trained for other tasks, cannot be assessed because appropriate psychometric methods are currently lacking. Here, we show how standard psychological questionnaires can be reformulated into natural language inference prompts, and we provide a code library to support the psychometric assessment of arbitrary models. We demonstrate, using a sample of 88 publicly available models, the existence of human-like mental health-related constructs (including anxiety, depression, and Sense of Coherence) which conform with standard theories in human psychology and show similar correlations and mitigation strategies. The ability to interpret and rectify the performance of language models by using psychological tools can boost the development of more explainable, controllable, and trustworthy models.

In an earlier paper, a randomized load balancing model was studied in a heavy traffic asymptotic regime where the load balancing stream is thin compared to the total arrival stream. It was shown that the limit is given by a system of rank-based Brownian particles on the half-line. This paper extends these results from the case of exponential service time to an invariance principle, where service times have finite second moment. The main tool is a new notion of rank-based stochastic differential inclusion, which may be of interest in its own right.

Machine/deep learning models have been widely adopted for predicting the configuration performance of software systems. However, a crucial yet unaddressed challenge is how to cater for the sparsity inherited from the configuration landscape: the influence of configuration options (features) and the distribution of data samples are highly sparse. In this paper, we propose a model-agnostic and sparsity-robust framework for predicting configuration performance, dubbed DaL, based on the new paradigm of dividable learning that builds a model via "divide-and-learn". To handle sample sparsity, the samples from the configuration landscape are divided into distant divisions, for each of which we build a sparse local model, e.g., regularized Hierarchical Interaction Neural Network, to deal with the feature sparsity. A newly given configuration would then be assigned to the right model of division for the final prediction. Further, DaL adaptively determines the optimal number of divisions required for a system and sample size without any extra training or profiling. Experiment results from 12 real-world systems and five sets of training data reveal that, compared with the state-of-the-art approaches, DaL performs no worse than the best counterpart on 44 out of 60 cases with up to 1.61x improvement on accuracy; requires fewer samples to reach the same/better accuracy; and producing acceptable training overhead. In particular, the mechanism that adapted the parameter d can reach the optimal value for 76.43% of the individual runs. The result also confirms that the paradigm of dividable learning is more suitable than other similar paradigms such as ensemble learning for predicting configuration performance. Practically, DaL considerably improves different global models when using them as the underlying local models, which further strengthens its flexibility., Comment: Submitted to TSE as a regular journal paper. arXiv admin note: text overlap with arXiv:2306.06651

In the Fully Sharded Data Parallel (FSDP) training pipeline, collective operations can be interleaved to maximize the communication/computation overlap. In this scenario, outstanding operations such as Allgather and Reduce-Scatter can compete for the injection bandwidth and create pipeline bubbles. To address this problem, we propose a novel bandwidth-optimal Allgather collective algorithm that leverages hardware multicast. We use multicast to build a constant-time reliable Broadcast protocol, a building block for constructing an optimal Allgather schedule. Our Allgather algorithm achieves 2x traffic reduction on a 188-node testbed. To free the host side from running the protocol, we employ SmartNIC offloading. We extract the parallelism in our Allgather algorithm and map it to a SmartNIC specialized for hiding the cost of data movement. We show that our SmartNIC-offloaded collective progress engine can scale to the next generation of 1.6 Tbit/s links.

Sadly, our esteemed colleague and friend Ning Cai passed away on 25th May, 2023. In his memory, Ingo Alth\"ofer, Holger Boche, Christian Deppe, Jens Stoye, Ulrich Tamm, Andreas Winter, and Raymond Yeung have organized the "Workshop on Information Theory and Related Fields" at the Bielefeld ZiF (Center for Interdisciplinary Research). This special event will be held from 24th November to 26th November, 2023. The workshop aims to celebrate Ning Cai's remarkable contributions to the field of information theory and to provide a platform for discussing current research in related areas. Ning Cai's work has had a significant impact on many domains, and this gathering will bring together colleagues, collaborators, and young researchers who have been influenced by his pioneering efforts.

Re-Identification systems (Re-ID) are crucial for public safety but face the challenge of having to adapt to environments that differ from their training distribution. Furthermore, rigorous privacy protocols in public places are being enforced as apprehensions regarding individual freedom rise, adding layers of complexity to the deployment of accurate Re-ID systems in new environments. For example, in the European Union, the principles of ``Data Minimization'' and ``Purpose Limitation'' restrict the retention and processing of images to what is strictly necessary. These regulations pose a challenge to the conventional Re-ID training schemes that rely on centralizing data on servers. In this work, we present a novel setting for privacy-preserving Distributed Unsupervised Domain Adaptation for person Re-ID (DUDA-Rid) to address the problem of domain shift without requiring any image transfer outside the camera devices. To address this setting, we introduce Fed-Protoid, a novel solution that adapts person Re-ID models directly within the edge devices. Our proposed solution employs prototypes derived from the source domain to align feature statistics within edge devices. Those source prototypes are distributed across the edge devices to minimize a distributed Maximum Mean Discrepancy (MMD) loss tailored for the DUDA-Rid setting. Our experiments provide compelling evidence that Fed-Protoid outperforms all evaluated methods in terms of both accuracy and communication efficiency, all while maintaining data privacy., Comment: ECCV 2024

Communication network engineering in enterprise environments is traditionally a complex, time-consuming, and error-prone manual process. Most research on network engineering automation has concentrated on configuration synthesis, often overlooking changes in the physical network topology. This paper introduces GeNet, a multimodal co-pilot for enterprise network engineers. GeNet is a novel framework that leverages a large language model (LLM) to streamline network design workflows. It uses visual and textual modalities to interpret and update network topologies and device configurations based on user intents. GeNet was evaluated on enterprise network scenarios adapted from Cisco certification exercises. Our results demonstrate GeNet's ability to interpret network topology images accurately, potentially reducing network engineers' efforts and accelerating network design processes in enterprise environments. Furthermore, we show the importance of precise topology understanding when handling intents that require modifications to the network's topology.

The observation of neutron stars enables the otherwise impossible study of fundamental physical processes. The timing of binary radio pulsars is particularly powerful, as it enables precise characterization of their (three-dimensional) positions and orbits. PSR~J0437$-$4715 is an important millisecond pulsar for timing array experiments and is also a primary target for the Neutron Star Interior Composition Explorer (NICER). The main aim of the NICER mission is to constrain the neutron star equation of state by inferring the compactness ($M_p/R$) of the star. Direct measurements of the mass $M_p$ from pulsar timing therefore substantially improve constraints on the radius $R$ and the equation of state. Here we use observations spanning 26 years from Murriyang, the 64-m Parkes radio telescope, to improve the timing model for this pulsar. Among the new precise measurements are the pulsar mass $M_p=1.418\pm 0.044$ $M_{\odot}$, distance $D=156.96 \pm 0.11$ pc, and orbital inclination angle $i=137.506 \pm 0.016^\circ$, which can be used to inform the X-ray pulse profile models inferred from NICER observations. We demonstrate that these results are consistent between multiple data sets from the Parkes Pulsar Timing Array (PPTA), each modeled with different noise assumptions. Using the longest available PPTA data set, we measure an apparent second derivative of the pulsar spin frequency and discuss how this can be explained either by kinematic effects due to the proper motion and radial velocity of the pulsar or excess low-frequency noise such as a gravitational-wave background., Comment: 13 pages, 3 figures, accepted for publication in Astrophysical Journal Letters

The flow of a $k$-cooperative system maps the set of vectors with up to~$(k-1)$ sign variations to itself. In particular, $1$-cooperative systems are just cooperative systems. Strongly $2$-cooperative systems satisfy a strong \Poincare-Bendixson property: any bounded solution that evolves in a compact set containing no equilibria converges to a periodic orbit. For $3$-dimensional strongly $2$-cooperative nonlinear systems, we provide a sufficient condition that guarantees the existence of an invariant compact set in the state space that includes an attracting periodic orbit. We show that our theoretical results unify and generalize known results on the existence of a periodic solution in two well-known models in biochemistry, a 3D Goodwin oscillator model and the 3D Field-Noyes ordinary-differential-equation (ODE) model for the Belousov-Zhabotinskii reaction, while simplifying the proofs.

This paper lays the groundwork for a unifying approach to stochastic dominance testing under survey nonresponse that integrates the partial identification approach to incomplete data and design-based inference for complex survey data. We propose a novel inference procedure for restricted $s$th-order stochastic dominance, tailored to accommodate a broad spectrum of nonresponse assumptions. The method uses pseudo-empirical likelihood to formulate the test statistic and compares it to a critical value from the chi-squared distribution with one degree of freedom. We detail the procedure's asymptotic properties under both null and alternative hypotheses, establishing its uniform validity under the null and consistency against various alternatives. Using the Household, Income and Labour Dynamics in Australia survey, we demonstrate the procedure's utility in a sensitivity analysis of temporal poverty comparisons among Australian households.

We introduce and analyze generalised polynomial chaos (gPC), considering both intrusive and non-intrusive approaches, as an uncertainty quantification method in studies of probabilistic robustness. The considered gPC methods are complementary to Monte Carlo (MC) methods and are shown to be fast and scalable, allowing for comprehensive and efficient exploration of parameter spaces. These properties enable robustness analysis of a wider set of models, compared to computationally expensive MC methods, while retaining desired levels of accuracy. We discuss the application of gPC methods to systems in biology and neuroscience, notably subject to multiple parametric uncertainties, and we examine a well-known model of neural dynamics as a case study.

Large Language Models (LLMs) frequently hallucinate, impeding their reliability in mission-critical situations. One approach to address this issue is to provide citations to relevant sources alongside generated content, enhancing the verifiability of generations. However, citing passages accurately in answers remains a substantial challenge. This paper proposes a weakly-supervised fine-tuning method leveraging factual consistency models (FCMs). Our approach alternates between generating texts with citations and supervised fine-tuning with FCM-filtered citation data. Focused learning is integrated into the objective, directing the fine-tuning process to emphasise the factual unit tokens, as measured by an FCM. Results on the ALCE few-shot citation benchmark with various instruction-tuned LLMs demonstrate superior performance compared to in-context learning, vanilla supervised fine-tuning, and state-of-the-art methods, with an average improvement of $34.1$, $15.5$, and $10.5$ citation F$_1$ points, respectively. Moreover, in a domain transfer setting we show that the obtained citation generation ability robustly transfers to unseen datasets. Notably, our citation improvements contribute to the lowest factual error rate across baselines., Comment: Accepted to ACL 2024. Code is available at https://github.com/amazon-science/learning-to-generate-answers-with-citations

In the rapidly evolving optimization and metaheuristics domains, the efficacy of algorithms is crucially determined by the benchmark (test) functions. While several functions have been developed and derived over the past decades, little information is available on the mathematical and visual description, range of suitability, and applications of many such functions. To bridge this knowledge gap, this review provides an exhaustive survey of more than 300 benchmark functions used in the evaluation of optimization and metaheuristics algorithms. This review first catalogs benchmark and test functions based on their characteristics, complexity, properties, visuals, and domain implications to offer a wide view that aids in selecting appropriate benchmarks for various algorithmic challenges. This review also lists the 25 most commonly used functions in the open literature and proposes two new, highly dimensional, dynamic and challenging functions that could be used for testing new algorithms. Finally, this review identifies gaps in current benchmarking practices and suggests directions for future research.

Shock waves driven by fast and wide coronal mass ejections (CMEs) are highly efficient particle accelerators involved in the production of solar energetic particle (SEP) events. The gradual SEP event measured by STEREO-A and B on October 11, 2013 had notable properties: (1) it occurred in isolation with very low background particle intensities, (2) it had a clear onset of SEPs measured in situ allowing detailed timing analyses, and (3) it was associated with a fast CME event magnetically connected with STA and B. These allowed us to investigate the temporal connection between the rapidly evolving shock properties, such as compression ratio, Mach number and geometry, and the intensity and composition of SEPs measured in situ. We use shock reconstruction techniques and multi-viewpoint imaging data from STA and B, SOHO, and SDO spacecraft to determine the kinematic evolution of the expanding shock wave. Using 3D magneto-hydrodynamic modelling we obtained shock wave properties along an ensemble of magnetic field lines connected to STA and B, estimating their uncertainties. Using a velocity dispersion analysis of the SEP data, we time shift the SEP time series and analyze the relations between their properties and the modeled shock ones, as well as the energy dependence of these relations. We find a very good temporal agreement between the formation of the modelled shock wave and the estimated release times for both electrons and protons. This simultaneous release suggests a common acceleration process. This early phase is marked at both STEREOs by elevated electron-to-proton ratios that coincide with the highly quasi-perpendicular phase of the shock, suggesting that the rapid evolution of the shock as it transits from the low to the high corona modifies the conditions under which particles are accelerated. We discuss these findings in terms of basic geometry and acceleration processes., Comment: 13 pages, 14 figures, submitted to A&A

The task of Visual Place Recognition (VPR) is to predict the location of a query image from a database of geo-tagged images. Recent studies in VPR have highlighted the significant advantage of employing pre-trained foundation models like DINOv2 for the VPR task. However, these models are often deemed inadequate for VPR without further fine-tuning on task-specific data. In this paper, we propose a simple yet powerful approach to better exploit the potential of a foundation model for VPR. We first demonstrate that features extracted from self-attention layers can serve as a powerful re-ranker for VPR. Utilizing these features in a zero-shot manner, our method surpasses previous zero-shot methods and achieves competitive results compared to supervised methods across multiple datasets. Subsequently, we demonstrate that a single-stage method leveraging internal ViT layers for pooling can generate global features that achieve state-of-the-art results, even when reduced to a dimensionality as low as 128D. Nevertheless, incorporating our local foundation features for re-ranking, expands this gap. Our approach further demonstrates remarkable robustness and generalization, achieving state-of-the-art results, with a significant gap, in challenging scenarios, involving occlusion, day-night variations, and seasonal changes.

Personalized retrieval and segmentation aim to locate specific instances within a dataset based on an input image and a short description of the reference instance. While supervised methods are effective, they require extensive labeled data for training. Recently, self-supervised foundation models have been introduced to these tasks showing comparable results to supervised methods. However, a significant flaw in these models is evident: they struggle to locate a desired instance when other instances within the same class are presented. In this paper, we explore text-to-image diffusion models for these tasks. Specifically, we propose a novel approach called PDM for Personalized Features Diffusion Matching, that leverages intermediate features of pre-trained text-to-image models for personalization tasks without any additional training. PDM demonstrates superior performance on popular retrieval and segmentation benchmarks, outperforming even supervised methods. We also highlight notable shortcomings in current instance and segmentation datasets and propose new benchmarks for these tasks., Comment: Accepted to NeurIPS 2024

Dynamic data-driven Digital Twins (DDTs) can enable informed decision-making and provide an optimisation platform for the underlying system. By leveraging principles of Dynamic Data-Driven Applications Systems (DDDAS), DDTs can formulate computational modalities for feedback loops, model updates and decision-making, including autonomous ones. However, understanding autonomous decision-making often requires technical and domain-specific knowledge. This paper explores using large language models (LLMs) to provide an explainability platform for DDTs, generating natural language explanations of the system's decision-making by leveraging domain-specific knowledge bases. A case study from smart agriculture is presented., Comment: 9 pages, 3 figures, accepted by DDDAS2024 -- the 5th International Conference on Dynamic Data Driven Applications Systems

In this paper, we first prove the existence of relative free models of morphisms (resp. relative commutative models) in the category of $DGA(R)$ (resp. $CDGA(R)$), where $R$ is a principal ideal domain containing $\frac{1}{2}$. Next, we restrict to the category of $(r,\rho(R))$-H-mild algebras and we introduce, following Carrasquel's characterization, $secat(-, R)$, the sectional category for surjective morphisms. We then apply this to the $n$-fold product of the commutative model of an $(r,\rho(R))$-mild CW-complex of finite type to introduce $TC_n(X,R)$, $mTC_n(X,R)$ and $HTC_n(X,R)$ which extend well known rational topological complexities. We do the same for $\operatorname{sc(-, \mathbb{Q})}$ to introduce analogous algebraic $\operatorname{sc(-,R)}$ in terms of their commutative models over $R$ and prove that it is an upper bound for $secat(-, R)$. This also yields, for any $(r,\rho(R))$-mild CW-complex, the algebraic $tc_n(X,R)$, $mtc_n(X,R)$ and $Htc_n(X,R)$ whose relation to the homology nilpotency is investigated. In the last section, in the same spirit, we introduce in $DGA(R)$, $secat(-, R)$, $\operatorname{sc(-,R)}$ and their topological correspondents. We then prove, in particular, that $ATC_n(X,R)\leq TC_n(X,R)$ and $Atc_n(X,R)\leq tc_n(X,R)$., Comment: 17 pages

We introduce a novel LLM based solution design approach that utilizes combinatorial optimization and sampling. Specifically, a set of factors that influence the quality of the solution are identified. They typically include factors that represent prompt types, LLM inputs alternatives, and parameters governing the generation and design alternatives. Identifying the factors that govern the LLM solution quality enables the infusion of subject matter expert knowledge. Next, a set of interactions between the factors are defined and combinatorial optimization is used to create a small subset $P$ that ensures all desired interactions occur in $P$. Each element $p \in P$ is then developed into an appropriate benchmark. Applying the alternative solutions on each combination, $p \in P$ and evaluating the results facilitate the design of a high quality LLM solution pipeline. The approach is especially applicable when the design and evaluation of each benchmark in $P$ is time-consuming and involves manual steps and human evaluation. Given its efficiency the approach can also be used as a baseline to compare and validate an autoML approach that searches over the factors governing the solution.

Collisionless shock waves, ubiquitous in the universe, are crucial for particle acceleration in various astrophysical systems. Currently, the heliosphere is the only natural environment available for their in situ study. In this work, we showcase the collective acceleration of electrons and ions by one of the fastest in situ shocks ever recorded, observed by the pioneering Parker Solar Probe at only 34.5 million kilometers from the Sun. Our analysis of this unprecedented, near-parallel shock shows electron acceleration up to 6 MeV amidst intense multi-scale electromagnetic wave emissions. We also present evidence of a variable shock structure capable of injecting and accelerating ions from the solar wind to high energies through a self-consistent process. The exceptional capability of the probe's instruments to measure electromagnetic fields in a shock traveling at 1% the speed of light has enabled us, for the first time, to confirm that the structure of a strong heliospheric shock aligns with theoretical models of strong shocks observed in astrophysical environments. This alignment offers viable avenues for understanding astrophysical shock processes and the acceleration of charged particles., Comment: Main text: 9 pages, 4 figures, 1 table; Supplementary text: 9 pages, 6 figures

We report the ASKAP discovery of a new Galactic supernova remnant (SNR) candidate G308.73+1.38, which we name Raspberry. This new SNR candidate has an angular size of 20.7 arcmin $\times$ 16.7 arcmin, and we measure a total integrated flux of 407$\pm$50 mJy. We estimate Raspberry's most likely diameter of 10$-$30 pc which would place it at a distance of 3$-$5 kpc, on the near side of the Milky Way's Scutum$-$Centaurus Arm. We also find a Stokes$-$V point source close to the centre of Raspberry with a $\sim$5$\sigma$ significance. This point source may be the remaining compact source, a neutron star, or possibly a pulsar, formed during the initial supernova event., Comment: 3 pages, 1 figure

We investigate the problem of common randomness (CR) generation in the basic two-party communication setting in which a sender and a receiver aim to agree on a common random variable with high probability. The terminals observe independent and identically distributed (i.i.d.) samples of sources with an arbitrary distribution defined on a Polish alphabet and are allowed to communicate as little as possible over a noisy, memoryless channel. We establish single-letter upper and lower bounds on the CR capacity for the specified model. The derived bounds hold with equality except for at most countably many points where discontinuity issues might arise., Comment: arXiv admin note: text overlap with arXiv:2210.04556

We assessed three schemes for propagating a variable-width (thawed) Gaussian wave packet moving under the influence of Morse or double-well potentials with parameters that are chemically representative. The most rigorous scheme is based on the time-dependent variational principle (TDVP); it leads to realistic behaviour of the center and width of a wave packet in all investigated regimes. Two other approximate schemes, Heller's and the extended semiclassical ones, demonstrate various aberrations. Heller's scheme does not properly account for various zero-point energy-related effects, is unable to predict tunneling, and more importantly, exhibits completely nonphysical unbound width oscillations. The extended semiclassical scheme, which was developed to address some of the shortcomings of the Heller counterpart, demonstrates another unphysical behaviour: self-trapping of a trajectory in both Morse and double-well potentials. We conclude that only the TDVP-based scheme is suitable for problem-free dynamical simulations. This, however, raises the question of how to utilize it efficiently in high-dimensional systems., Comment: 11 pages, 8 figures

A method for performing variable-width (thawed) Gaussian wavepacket (GWP) variational dynamics on machine-learned potentials is presented. Instead of fitting the potential energy surface (PES), the anharmonic correction to the global harmonic approximation (GHA) is fitted using kernel ridge regression -- this is a $\Delta$-machine learning approach. The training set consists of energy differences between ab initio electronic energies and values given by the GHA. The learned potential is subsequently used to propagate a single thawed GWP using the time-dependent variational principle to compute the autocorrelation function, which provides direct access to vibronic spectra via its Fourier transform. We applied the developed method to simulate the photoelectron spectrum of ammonia and found excellent agreement between theoretical and experimental spectra. We show that fitting the anharmonic corrections requires a smaller training set as compared to fitting total electronic energies. We also demonstrate that our approach allows to reduce the dimensionality of the nuclear space used to scan the PES when constructing the training set. Thus, only the degrees of freedom associated with large amplitude motions need to be treated with $\Delta$-machine learning, which paves a way for reliable simulations of vibronic spectra of large floppy molecules., Comment: 12 pages, 3 figures

Many existing motion prediction approaches rely on symbolic perception outputs to generate agent trajectories, such as bounding boxes, road graph information and traffic lights. This symbolic representation is a high-level abstraction of the real world, which may render the motion prediction model vulnerable to perception errors (e.g., failures in detecting open-vocabulary obstacles) while missing salient information from the scene context (e.g., poor road conditions). An alternative paradigm is end-to-end learning from raw sensors. However, this approach suffers from the lack of interpretability and requires significantly more training resources. In this work, we propose tokenizing the visual world into a compact set of scene elements and then leveraging pre-trained image foundation models and LiDAR neural networks to encode all the scene elements in an open-vocabulary manner. The image foundation model enables our scene tokens to encode the general knowledge of the open world while the LiDAR neural network encodes geometry information. Our proposed representation can efficiently encode the multi-frame multi-modality observations with a few hundred tokens and is compatible with most transformer-based architectures. To evaluate our method, we have augmented Waymo Open Motion Dataset with camera embeddings. Experiments over Waymo Open Motion Dataset show that our approach leads to significant performance improvements over the state-of-the-art., Comment: CVPR 2024

This work presents mm-Wave and sub-THz chip-to-package transitions for communications systems. To date, reported transitions either have high loss, typically 3 to 4 dB, or require high cost packages to support very fine bump pitches and low loss materials. We analyze the impact of transitions on a high frequency, wide bandwidth communication system and present the design of a chip-to-package transition in two different commercial packaging technologies. The proposed transitions achieve <1 dB loss in both technologies, validating the design methodology., Comment: 8 pages, 16 figures, to be submitted to an IEEE Journal. The first two authors are co-first authors

With the rise of satellites and mankind's growing dependence on technology, there is an increasing awareness of space weather phenomena related to high-energy particles. Shock waves driven by coronal mass ejections (CMEs) and corotating interaction regions (CIRs) occasionally act as potent particle accelerators, generating hazardous solar energetic particles (SEPs) that pose risks to satellite electronics and astronauts. Numerical simulation tools capable of modelling and predicting large SEP events are thus highly demanded. We introduce the new Icarus$+$PARADISE model as an advancement of the previous EUHFORIA$+$PARADISE model. Icarus, based on the MPI-AMRVAC framework, is a three-dimensional magnetohydrodynamic code that models solar wind configurations from 0.1 au onwards, encompassing transient structures like CMEs or CIRs. Differing from EUHFORIA's uniform-only grid, Icarus incorporates solution adaptive mesh refinement (AMR) and grid stretching. The particle transport code PARADISE propagates energetic particles as test particles through these solar wind configurations by solving the focused transport equation in a stochastic manner. We validate our new model by reproducing EUHFORIA+PARADISE results. This is done by modelling the acceleration and transport of energetic particles in a synthetic solar wind configuration containing an embedded CIR. Subsequently, we illustrate how the simulation results vary with grid resolution by employing different levels of AMR. The resulting intensity profiles illustrate increased particle acceleration with higher levels of AMR in the shock region, better capturing the effects of the shock., Comment: 20 pages, 8 figures

This paper describes AssemblyAI's industrial-scale automatic speech recognition (ASR) system, designed to meet the requirements of large-scale, multilingual ASR serving various application needs. Our system leverages a diverse training dataset comprising unsupervised (12.5M hours), supervised (188k hours), and pseudo-labeled (1.6M hours) data across four languages. We provide a detailed description of our model architecture, consisting of a full-context 600M-parameter Conformer encoder pre-trained with BEST-RQ and an RNN-T decoder fine-tuned jointly with the encoder. Our extensive evaluation demonstrates competitive word error rates (WERs) against larger and more computationally expensive models, such as Whisper large and Canary-1B. Furthermore, our architectural choices yield several key advantages, including an improved code-switching capability, a 5x inference speedup compared to an optimized Whisper baseline, a 30% reduction in hallucination rate on speech data, and a 90% reduction in ambient noise compared to Whisper, along with significantly improved time-stamp accuracy. Throughout this work, we adopt a system-centric approach to analyzing various aspects of fully-fledged ASR models to gain practically relevant insights useful for real-world services operating at scale.

Given a reaction-diffusion equation with unknown right-hand side, we consider a nonlinear inverse problem of estimating the associated leading eigenvalues and initial condition modes from a finite number of non-local noisy measurements. We define a reconstruction criterion and, for a small enough noise, we prove the existence and uniqueness of the desired approximation and derive closed-form expressions for the first-order condition numbers, as well as bounds for their asymptotic behavior in a regime when the number of measured samples is fixed and the inter-sampling interval length tends to infinity. When computing the sought estimates numerically, our simulations show that the exponential fitting algorithm ESPRIT is first-order optimal, as its first-order condition numbers have the same asymptotic behavior as the analytic condition numbers in the considered regime.

This study investigates failure mechanisms in a typical thermal barrier coating (TBC) system comprising an EB-PVD columnar top coat, an aluminide bond coat, and a Ni-based single crystal superalloy substrate, simulating gas turbine operating conditions using a burner rig. TBC degradation, initiated by interfacial defects from the LASAT method, was studied during thermal gradient cycling under fast and slow cooling. In-situ optical and infrared imaging, along with ex-situ SEM cross-sectional analysis, monitored failure mechanisms. The Laser Shock for Damage Monitoring (LASDAM) technique provided insights into gradient and cooling rate impacts on columnar TBC damage. Results showed significant effects of cooling rate on delamination and localized failure at blister sites, with LASDAM revealing significant overheating at damage sites. Analysis included full-field temperature and damage assessment, emphasizing blister-driven delamination under severe thermal gradients. Discussion focused on elastic stored energy effects, noting that fast cooling induced transient conditions where reversed temperature gradients increased damage, limiting TBC lifespan, Comment: submitted to Journal of the European Ceramic Society, 24 pages and 19 figures

Motion forecasting has become an increasingly critical component of autonomous robotic systems. Onboard compute budgets typically limit the accuracy of real-time systems. In this work we propose methods of improving motion forecasting systems subject to limited compute budgets by combining model ensemble and distillation techniques. The use of ensembles of deep neural networks has been shown to improve generalization accuracy in many application domains. We first demonstrate significant performance gains by creating a large ensemble of optimized single models. We then develop a generalized framework to distill motion forecasting model ensembles into small student models which retain high performance with a fraction of the computing cost. For this study we focus on the task of motion forecasting using real world data from autonomous driving systems. We develop ensemble models that are very competitive on the Waymo Open Motion Dataset (WOMD) and Argoverse leaderboards. From these ensembles, we train distilled student models which have high performance at a fraction of the compute costs. These experiments demonstrate distillation from ensembles as an effective method for improving accuracy of predictive models for robotic systems with limited compute budgets., Comment: 11 pages, 14 figures

This paper introduces PRobELM (Plausibility Ranking Evaluation for Language Models), a benchmark designed to assess language models' ability to discern more plausible from less plausible scenarios through their parametric knowledge. While benchmarks such as TruthfulQA emphasise factual accuracy or truthfulness, and others such as COPA explore plausible scenarios without explicitly incorporating world knowledge, PRobELM seeks to bridge this gap by evaluating models' capabilities to prioritise plausible scenarios that leverage world knowledge over less plausible alternatives. This design allows us to assess the potential of language models for downstream use cases such as literature-based discovery where the focus is on identifying information that is likely but not yet known. Our benchmark is constructed from a dataset curated from Wikidata edit histories, tailored to align the temporal bounds of the training data for the evaluated models. PRobELM facilitates the evaluation of language models across multiple prompting types, including statement, text completion, and question-answering. Experiments with 10 models of various sizes and architectures on the relationship between model scales, training recency, and plausibility performance, reveal that factual accuracy does not directly correlate with plausibility performance and that up-to-date training data enhances plausibility assessment across different model architectures.

Inspired by the picture portraying the KNO scaling violation as an extension of the geometrical scaling violation, the current study proposes a phenomenological model for multi-particle production in hadron collisions based on the geometrical approach and using the $U$-Matrix unitarization scheme of the scattering amplitude. The model has been fine-tuned and all parameters have been derived from optimal fits to various hadronic multiplicity distributions data in $p + p(\bar{p})$ collisions across a broad range of energies. The results have revealed that our model furnishes a reasonable description of diverse multiplicity distributions at various energies. Besides, they have demonstrated a pronounced violation of the geometrical scaling, which eventually resulted in a significant violation of the KNO scaling. The study has also analyzed the higher-order moments of the multiplicity distribution. We have observed an unexpected overestimation of the fluctuations and correlations between final state particles with increasing energy, particularly above LHC energy. It is claimed that this overestimation is due to statistical fluctuations embedded in the $U$-matrix scheme. The findings of this study have shed light on the key role of the $U$-matrix scheme in the impact of collision geometry on multi-particle production processes at high energy.

We consider a continuous-time random walk in the quarter plane for which the transition intensities are constant on each of the four faces $(0,\infty)^2$, $F_1=\{0\}\times(0,\infty)$, $F_2=(0,\infty)\times\{0\}$ and $\{(0,0)\}$. We show that when rescaled diffusively it converges in law to a Brownian motion with oblique reflection direction $d^{(i)}$ on face $F_i$, $i=1,2$, defined via the Varadhan-Williams submartingale problem. A parameter denoted by $\alpha$ was introduced in \cite{vw}, measuring the extent to which $d^{(i)}$ are inclined toward the origin. In the case of the quarter plane, $\alpha$ takes values in $(-2,2)$, and it is known that the reflected Brownian motion is a semimartingale if and only if $\alpha\in(-2,1)$. Convergence results via both the Skorohod map and the invariance principle for semimartingale reflected Brownian motion are known to hold in various settings in arbitrary dimension. In the case of the quarter plane, the invariance principle was proved for $\alpha \in (-2,1)$ whereas for tools based on the Skorohod map to be applicable it is necessary (but not sufficient) that $\alpha \in [-1,1)$. Another tool that has been used to prove convergence in general dimension is the extended Skorohod map, which in the case of the quarter plane provides convergence for $\alpha=1$. This paper focuses on the range $\alpha \in (1,2)$, where the Skorohod problem and the extended Skorohod problem do not possess a unique solution, the limit process is not a semimartingale, and convergence to reflected Brownian motion has not been shown before. The result has implications on the asymptotic analysis of two Markovian queueing models: The {\it generalized processor sharing model with parallelization slowdown}, and the {\it coupled processor model}.

We consider a load balancing model where a Poisson stream of jobs arrive at a system of many servers whose service time distribution possesses a finite second moment. A small fraction of arrivals pass through the so called power-of-choice algorithm, which assigns a job to the shortest among $\ell$, $\ell\ge 2$, randomly chosen queues, and the remaining jobs are assigned to queues chosen uniformly at random. The system is analyzed at critical load in an asymptotic regime where both the number of servers and the usual heavy traffic parameter associated with individual queue lengths grow to infinity. The first main result is a hydrodynamic limit, where the empirical measure of the diffusively normalized queue lengths is shown to converge to a path in measure space whose density is given by the unique solution of a parabolic PDE with nonlocal coefficients. Further, two forms of an invariance principle are proved, corresponding to two different assumptions on the initial distribution, where individual normalized queue lengths converge weakly to solutions of SDE. In one of these results, the limit is given by a McKean-Vlasov SDE, and propagation of chaos holds. The McKean-Vlasov limit is closely related to limit results for Brownian particles on $\mathbb{R}_+$ interacting through their rank (with a specific interaction). However, an entirely different set of tools is required, as the collection of $n$ prelimit particles does not obey a Markovian evolution on $\mathbb{R}_+^n$.

Online Unsupervised Domain Adaptation (OUDA) for person Re-Identification (Re-ID) is the task of continuously adapting a model trained on a well-annotated source domain dataset to a target domain observed as a data stream. In OUDA, person Re-ID models face two main challenges: catastrophic forgetting and domain shift. In this work, we propose a new Source-guided Similarity Preservation (S2P) framework to alleviate these two problems. Our framework is based on the extraction of a support set composed of source images that maximizes the similarity with the target data. This support set is used to identify feature similarities that must be preserved during the learning process. S2P can incorporate multiple existing UDA methods to mitigate catastrophic forgetting. Our experiments show that S2P outperforms previous state-of-the-art methods on multiple real-to-real and synthetic-to-real challenging OUDA benchmarks., Comment: WACV 2024

Security by design, Sbd is a concept for developing and maintaining systems that are, to the greatest extent possible, free from security vulnerabilities and impervious to security attacks. In addition to technical aspects, such as how to develop a robust industrial control systems hardware, software, communication product, etc., SbD includes also soft aspects, such as organizational managerial attitude and behavior, and employee awareness. Under the Sbd concept, systems, ICS in our context, will be considered more trustworthy by users. User's trust in the systems will be derived from the meticulous adherence to the SbD processes and policies. In accordance with the SbD concept, security is considered. Security measures are implemented, at every stage of the product and systems development life cycle, rather than afterwards. This document presents the security requirements for the implementation of the SbD in industrial control systems. The information presented does not negate any existing security and cyber security standards, etc. Instead, we strongly recommend that organizations should implement and comply with those standards and best practices. Security by design is not a one-time process. It starts at the very beginning of the products of the system design and continues through all its lifecycle. Due to the benefits of the SbD, higher level of security, and robustness to cyber attacks, all organizations associated with the energy sector should strive to establish an ecosystem. The requirements presented in this document may be perceived as burdensome by organizations. However, strict compliance with the requirements and existing security standards and best practices, including continuous monitoring, as specified in this document, is essential to realize an ecosystem driven and protected by the SbD, Comment: 66 pages, 2 figures

How can we best encode structured data into sequential form for use in large language models (LLMs)? In this work, we introduce a parameter-efficient method to explicitly represent structured data for LLMs. Our method, GraphToken, learns an encoding function to extend prompts with explicit structured information. Unlike other work which focuses on limited domains (e.g. knowledge graph representation), our work is the first effort focused on the general encoding of structured data to be used for various reasoning tasks. We show that explicitly representing the graph structure allows significant improvements to graph reasoning tasks. Specifically, we see across the board improvements - up to 73% points - on node, edge and, graph-level tasks from the GraphQA benchmark.

This study assesses deep learning models for audio classification in a clinical setting with the constraint of small datasets reflecting real-world prospective data collection. We analyze CNNs, including DenseNet and ConvNeXt, alongside transformer models like ViT, SWIN, and AST, and compare them against pre-trained audio models such as YAMNet and VGGish. Our method highlights the benefits of pre-training on large datasets before fine-tuning on specific clinical data. We prospectively collected two first-of-their-kind patient audio datasets from stroke patients. We investigated various preprocessing techniques, finding that RGB and grayscale spectrogram transformations affect model performance differently based on the priors they learn from pre-training. Our findings indicate CNNs can match or exceed transformer models in small dataset contexts, with DenseNet-Contrastive and AST models showing notable performance. This study highlights the significance of incremental marginal gains through model selection, pre-training, and preprocessing in sound classification; this offers valuable insights for clinical diagnostics that rely on audio classification., Comment: CHIL 2024

We investigate the problem of generating common randomness (CR) from finite compound sources aided by unidirectional communication over rate-limited perfect channels. The two communicating parties, often referred to as terminals, observe independent and identically distributed (i.i.d.) samples of a finite compound source and aim to agree on a common random variable with a high probability for every possible realization of the source state. Both parties know the set of source states as well as their statistics. However, they are unaware of the actual realization of the source state. We establish a single-letter lower and upper bound on the compound CR capacity for the specified model. Furthermore, we present two special scenarios where the established bounds coincide., Comment: arXiv admin note: text overlap with arXiv:2305.05524

Efficiently generating energetically stable crystal structures has long been a challenge in material design, primarily due to the immense arrangement of atoms in a crystal lattice. To facilitate the discovery of stable material, we present a framework for the generation of synthesizable materials, leveraging a point cloud representation to encode intricate structural information. At the heart of this framework lies the introduction of a diffusion model as its foundational pillar. To gauge the efficacy of our approach, we employ it to reconstruct input structures from our training datasets, rigorously validating its high reconstruction performance. Furthermore, we demonstrate the profound potential of Point Cloud-Based Crystal Diffusion (PCCD) by generating entirely new materials, emphasizing their synthesizability. Our research stands as a noteworthy contribution to the advancement of materials design and synthesis through the cutting-edge avenue of generative design instead of the conventional substitution or experience-based discovery., Comment: I have submitted to a journal

The Circle of Willis (CoW) is an important network of arteries connecting major circulations of the brain. Its vascular architecture is believed to affect the risk, severity, and clinical outcome of serious neuro-vascular diseases. However, characterizing the highly variable CoW anatomy is still a manual and time-consuming expert task. The CoW is usually imaged by two angiographic imaging modalities, magnetic resonance angiography (MRA) and computed tomography angiography (CTA), but there exist limited public datasets with annotations on CoW anatomy, especially for CTA. Therefore we organized the TopCoW Challenge in 2023 with the release of an annotated CoW dataset. The TopCoW dataset was the first public dataset with voxel-level annotations for thirteen possible CoW vessel components, enabled by virtual-reality (VR) technology. It was also the first large dataset with paired MRA and CTA from the same patients. TopCoW challenge formalized the CoW characterization problem as a multiclass anatomical segmentation task with an emphasis on topological metrics. We invited submissions worldwide for the CoW segmentation task, which attracted over 140 registered participants from four continents. The top performing teams managed to segment many CoW components to Dice scores around 90%, but with lower scores for communicating arteries and rare variants. There were also topological mistakes for predictions with high Dice scores. Additional topological analysis revealed further areas for improvement in detecting certain CoW components and matching CoW variant topology accurately. TopCoW represented a first attempt at benchmarking the CoW anatomical segmentation task for MRA and CTA, both morphologically and topologically., Comment: 24 pages, 11 figures, 9 tables. Summary Paper for the MICCAI TopCoW 2023 Challenge