Search

Research organisms provide invaluable insights into human biology and diseases, serving as essential tools for functional experiments, disease modeling, and drug testing. However, evolutionary divergence between humans and research organisms hinders effective knowledge transfer across species. Here, we review state-of-the-art methods for computationally transferring knowledge across species, primarily focusing on methods that utilize transcriptome data and/or molecular networks. We introduce the term "agnology" to describe the functional equivalence of molecular components regardless of evolutionary origin, as this concept is becoming pervasive in integrative data-driven models where the role of evolutionary origin can become unclear. Our review addresses four key areas of information and knowledge transfer across species: (1) transferring disease and gene annotation knowledge, (2) identifying agnologous molecular components, (3) inferring equivalent perturbed genes or gene sets, and (4) identifying agnologous cell types. We conclude with an outlook on future directions and several key challenges that remain in cross-species knowledge transfer.

The momentum ray transform $I_m^k$ integrates a rank $m$ symmetric tensor field $f$ on ${\mathbb R}^n$ over lines with the weight $t^k$, $I_m^kf(x,\xi)=\int_{-\infty}^\infty t^k\langle f(x+t\xi),\xi^m\rangle\,\mathrm{d}t$. Let $N^k_m=(I^k_m)^*I^k_m$ be the normal operator of $I_m^k$. To what extent is a symmetric $m$-tensor field $f$ determined by the data $(N_m^0f,\dots,N_m^rf)$ given for some $0\le r\le m$? The Saint Venant operator $W^r_m$ is a linear differential operator of order $m-r$ with constant coefficients on the space of symmetric $m$-tensor fields. We derive an explicit formula expressing $W^r_mf$ in terms of $(N_m^0f,\dots,N_m^rf)$. The tensor field $W^r_mf$ represents the full local information on $f$ that can be extracted from the data $(N_m^0f,\dots,N_m^rf)$., Comment: 10 pages. arXiv admin note: text overlap with arXiv:2401.00791

Liver allograft failure occurs in approximately 20% of liver transplant recipients within five years post-transplant, leading to mortality or the need for retransplantation. Providing an accurate and interpretable model for individualized risk estimation of graft failure is essential for improving post-transplant care. To this end, we introduce the Model for Allograft Survival (MAS), a simple linear risk score that outperforms other advanced survival models. Using longitudinal patient follow-up data from the United States (U.S.), we develop our models on 82,959 liver transplant recipients and conduct multi-site evaluations on 11 regions. Additionally, by testing on a separate non-U.S. cohort, we explore the out-of-distribution generalization performance of various models without additional fine-tuning, a crucial property for clinical deployment. We find that the most complex models are also the ones most vulnerable to distribution shifts despite achieving the best in-distribution performance. Our findings not only provide a strong risk score for predicting long-term graft failure but also suggest that the routine machine learning pipeline with only in-distribution held-out validation could create harmful consequences for patients at deployment., Comment: Accepted at MLHC 2024

The Large Magellanic Cloud (LMC) presents a unique environment for pulsar population studies due to its distinct star formation characteristics and proximity to the Milky Way. As part of the TRAPUM (TRAnsients and PUlsars with MeerKAT) Large Survey Project, we are using the core array of the MeerKAT radio telescope (MeerKAT) to conduct a targeted search of the LMC for radio pulsars at L-band frequencies, 856-1712$\,$MHz. The excellent sensitivity of MeerKAT, coupled with a 2-hour integration time, makes the survey 3 times more sensitive than previous LMC radio pulsar surveys. We report the results from the initial four survey pointings which has resulted in the discovery of seven new radio pulsars, increasing the LMC radio pulsar population by 30 per cent. The pulse periods of these new pulsars range from 278 to 1690$\,$ms, and the highest dispersion measure is 254.20$\,$pc$\,$cm$^{-3}$. We searched for, but did not find any significant pulsed radio emission in a beam centred on the SN$\,$1987A remnant, establishing an upper limit of 6.3$\,{\mu}$Jy on its minimum flux density at 1400$\,$MHz., Comment: 12 pages, 4 figures, 4 tables. Accepted for publication in Monthly Notices of the Royal Astronomical Society

We compute the second moment of the Riemann zeta function for shifted arguments over a domain that extends the ones in the literature. We use the Riemann-Siegel formula for the error term in the approximate functional equation and take the products of all the terms into account. We also show that, as a function of imaginary shifts on the critical line, the the second moment behaves like a Fourier-Cauchy type kernel on a class of functions. This is reminiscent of orthogonal functions., Comment: 17 pages

We report new radio timing solutions from a three-year observing campaign conducted with the MeerKAT and Murriyang telescopes for nine Small Magellanic Cloud pulsars, increasing the number of characterised rotation-powered extragalactic pulsars by 40 per cent. We can infer from our determined parameters that the pulsars are seemingly all isolated, that six are ordinary pulsars, and that three of the recent MeerKAT discoveries have a young characteristic age of under 100 kyr and have undergone a spin-up glitch. Two of the sources, PSRs J0040$-$7337 and J0048$-$7317, are energetic young pulsars with spin-down luminosities of the order of 10$^{36}$ erg s$^{-1}$. They both experienced a large glitch, with a change in frequency of about 30 $\mu$Hz, and a frequency derivative change of order $-10^{-14}$ Hz s$^{-1}$. These glitches, the inferred glitch rate, and the properties of these pulsars (including potentially high inter-glitch braking indices) suggest these neutron stars might be Vela-like repeating glitchers and should be closely monitored in the future. The position and energetics of PSR J0048$-$7317 confirm it is powering a new Pulsar Wind Nebula (PWN) detected as a radio continuum source; and similarly the association of PSR J0040$-$7337 with the PWN of Supernova Remnant (SNR) DEM S5 (for which we present a new Chandra image) is strengthened. Finally, PSR J0040$-$7335 is also contained within the same SNR but is a chance superposition. It has also been seen to glitch with a change of frequency of $10^{-2}$ $\mu$Hz. This work more than doubles the characterised population of SMC radio pulsars., Comment: 20 pages, 13 figures, accepted for publication in Monthly Notices of the Royal Astronomical Society

To date, there has been no comprehensive study characterizing the effect of diffusion-weighted magnetic resonance imaging voxel resolution on the resulting connectome for high resolution subject data. Similarity in results improved with higher resolution, even after initial down-sampling. To ensure robust tractography and connectomes, resample data to 1 mm isotropic resolution.

Modern artificial intelligence (AI) systems are powered by foundation models. This paper presents a new set of foundation models, called Llama 3. It is a herd of language models that natively support multilinguality, coding, reasoning, and tool usage. Our largest model is a dense Transformer with 405B parameters and a context window of up to 128K tokens. This paper presents an extensive empirical evaluation of Llama 3. We find that Llama 3 delivers comparable quality to leading language models such as GPT-4 on a plethora of tasks. We publicly release Llama 3, including pre-trained and post-trained versions of the 405B parameter language model and our Llama Guard 3 model for input and output safety. The paper also presents the results of experiments in which we integrate image, video, and speech capabilities into Llama 3 via a compositional approach. We observe this approach performs competitively with the state-of-the-art on image, video, and speech recognition tasks. The resulting models are not yet being broadly released as they are still under development.

Anomaly detection in computational workflows is critical for ensuring system reliability and security. However, traditional rule-based methods struggle to detect novel anomalies. This paper leverages large language models (LLMs) for workflow anomaly detection by exploiting their ability to learn complex data patterns. Two approaches are investigated: 1) supervised fine-tuning (SFT), where pre-trained LLMs are fine-tuned on labeled data for sentence classification to identify anomalies, and 2) in-context learning (ICL) where prompts containing task descriptions and examples guide LLMs in few-shot anomaly detection without fine-tuning. The paper evaluates the performance, efficiency, generalization of SFT models, and explores zero-shot and few-shot ICL prompts and interpretability enhancement via chain-of-thought prompting. Experiments across multiple workflow datasets demonstrate the promising potential of LLMs for effective anomaly detection in complex executions., Comment: 12 pages, 14 figures, paper is accepted by SC'24, source code, see: https://github.com/PoSeiDon-Workflows/LLM_AD

Semantic segmentation has emerged as a pivotal area of study in computer vision, offering profound implications for scene understanding and elevating human-machine interactions across various domains. While 2D semantic segmentation has witnessed significant strides in the form of lightweight, high-precision models, transitioning to 3D semantic segmentation poses distinct challenges. Our research focuses on achieving efficiency and lightweight design for 3D semantic segmentation models, similar to those achieved for 2D models. Such a design impacts applications of 3D semantic segmentation where memory and latency are of concern. This paper introduces a novel approach to 3D semantic segmentation, distinguished by incorporating a hybrid blend of 2D and 3D computer vision techniques, enabling a streamlined, efficient process. We conduct 2D semantic segmentation on RGB images linked to 3D point clouds and extend the results to 3D using an extrusion technique for specific class labels, reducing the point cloud subspace. We perform rigorous evaluations with the DeepViewAgg model on the complete point cloud as our baseline by measuring the Intersection over Union (IoU) accuracy, inference time latency, and memory consumption. This model serves as the current state-of-the-art 3D semantic segmentation model on the KITTI-360 dataset. We can achieve heightened accuracy outcomes, surpassing the baseline for 6 out of the 15 classes while maintaining a marginal 1% deviation below the baseline for the remaining class labels. Our segmentation approach demonstrates a 1.347x speedup and about a 43% reduced memory usage compared to the baseline., Comment: 18 pages, 3 figures, 3 tables

We introduce FedDM, a novel training framework designed for the federated training of diffusion models. Our theoretical analysis establishes the convergence of diffusion models when trained in a federated setting, presenting the specific conditions under which this convergence is guaranteed. We propose a suite of training algorithms that leverage the U-Net architecture as the backbone for our diffusion models. These include a basic Federated Averaging variant, FedDM-vanilla, FedDM-prox to handle data heterogeneity among clients, and FedDM-quant, which incorporates a quantization module to reduce the model update size, thereby enhancing communication efficiency across the federated network. We evaluate our algorithms on FashionMNIST (28x28 resolution), CIFAR-10 (32x32 resolution), and CelebA (64x64 resolution) for DDPMs, as well as LSUN Church Outdoors (256x256 resolution) for LDMs, focusing exclusively on the imaging modality. Our evaluation results demonstrate that FedDM algorithms maintain high generation quality across image resolutions. At the same time, the use of quantized updates and proximal terms in the local training objective significantly enhances communication efficiency (up to 4x) and model convergence, particularly in non-IID data settings, at the cost of increased FID scores (up to 1.75x)., Comment: 13 pages,3 figures, 2 algorithms, 3 tables

PSR J1227$-$6208 is a 34.53-ms recycled pulsar with a massive companion. This system has long been suspected to belong to the emerging class of massive recycled pulsar-ONeMg white dwarf systems such as PSR J2222$-$0137, PSR J1528$-$3146 and J1439$-$5501. Here we present an updated emission and timing analysis with more than 11 years of combined Parkes and MeerKAT data, including 19 hours of high-frequency data from the newly installed MeerKAT S-band receivers. We measure a scattering timescale of 1.22 ms at 1 GHz with a flat scattering index 3.33<$\beta$<3.62, and a mean flux density of 0.53-0.62 mJy at 1 GHz with a steep spectral index 2.06<$\alpha$<2.35. Around 15% of the emission is linearly and circularly polarised, but the polarisation angle does not follow the rotating vector model. Thanks to the sensitivity of MeerKAT, we successfully measure a rate of periastron advance of 0.0171(11) deg/yr, and a Shapiro delay with an orthometric amplitude of 3.6$\pm$0.5 $\mu$s and an orthometric shape of 0.85$\pm$0.05. The main source of uncertainty in our timing analysis is chromatic correlated dispersion measure noise, which we model as a power law in the Fourier space thanks to the large frequency coverage provided by the Parkes UWL receiver. Assuming general relativity and accounting for the measurements across all the implemented timing noise models, the total mass, companion mass, pulsar mass and inclination angle are constrained at 2.3

It has been suggested that black hole microstates in string theory may have "no interiors". The arguments of arXiv:2312.14108 suggest that even if this were the case, conventional bulk EFT should be valid at the horizon for black holes that are not too old. Here, we explore the validity of bulk EFT for young black holes from a different perspective, by considering the gravitational collapse of scalar moduli. This builds on arXiv:2003.05488, which studied the numerical collapse of a scalar field in a Choptuik-like set up. Evidence was presented there for two observations -- (a) For every scalar profile that results in collapse, the scalar field undergoes trans-Planckian variation, (b) Every trans-Planckian scalar motion is hidden behind an apparent horizon. The results of arXiv:2003.05488 were mostly obtained in regimes close to critical collapse, where scale-separation was difficult to achieve. In this paper, we study the same system in super-critical regimes. This allows us to study the localization properties of scalar variations inside the apparent horizon by separating the horizon and singularity scales. We find strong evidence that the $\mathcal{O}(1)$ scalar variation is localized around the curvature divergence, and is hierarchically well-separated from the apparent horizon. Because $\mathcal{O}(1)$ scalar movement is associated to the breakdown of bulk EFT according to the swampland distance conjecture, this is an indication that bulk EFT is well-behaved in the interiors of young black holes. Our results suggest the perspective that cosmic censorship is a mechanism for preserving bulk EFT between the horizon and the singularity., Comment: 16 pages, 16 plots

Artificial intelligence and machine learning frameworks have served as computationally efficient mapping between inputs and outputs for engineering problems. These mappings have enabled optimization and analysis routines that have warranted superior designs, ingenious material systems and optimized manufacturing processes. A common occurrence in such modeling endeavors is the existence of multiple source of data, each differentiated by fidelity, operating conditions, experimental conditions, and more. Data fusion frameworks have opened the possibility of combining such differentiated sources into single unified models, enabling improved accuracy and knowledge transfer. However, these frameworks encounter limitations when the different sources are heterogeneous in nature, i.e., not sharing the same input parameter space. These heterogeneous input scenarios can occur when the domains differentiated by complexity, scale, and fidelity require different parametrizations. Towards addressing this void, a heterogeneous multi-source data fusion framework is proposed based on input mapping calibration (IMC) and latent variable Gaussian process (LVGP). In the first stage, the IMC algorithm is utilized to transform the heterogeneous input parameter spaces into a unified reference parameter space. In the second stage, a multi-source data fusion model enabled by LVGP is leveraged to build a single source-aware surrogate model on the transformed reference space. The proposed framework is demonstrated and analyzed on three engineering case studies (design of cantilever beam, design of ellipsoidal void and modeling properties of Ti6Al4V alloy). The results indicate that the proposed framework provides improved predictive accuracy over a single source model and transformed but source unaware model., Comment: 20 Pages,9 Figures, Data is available per request

When Large Language Models (LLMs) are compressed using techniques such as quantization, the predominant way to demonstrate the validity of such techniques is by measuring the model's accuracy on various benchmarks.If the accuracies of the baseline model and the compressed model are close, it is assumed that there was negligible degradation in quality.However, even when the accuracy of baseline and compressed model are similar, we observe the phenomenon of flips, wherein answers change from correct to incorrect and vice versa in proportion.We conduct a detailed study of metrics across multiple compression techniques, models and datasets, demonstrating that the behavior of compressed models as visible to end-users is often significantly different from the baseline model, even when accuracy is similar.We further evaluate compressed models qualitatively and quantitatively using MT-Bench and show that compressed models are significantly worse than baseline models in this free-form generative task.Thus, we argue that compression techniques should also be evaluated using distance metrics.We propose two such metrics, KL-Divergence and flips, and show that they are well correlated.

We propose OmniNOCS, a large-scale monocular dataset with 3D Normalized Object Coordinate Space (NOCS) maps, object masks, and 3D bounding box annotations for indoor and outdoor scenes. OmniNOCS has 20 times more object classes and 200 times more instances than existing NOCS datasets (NOCS-Real275, Wild6D). We use OmniNOCS to train a novel, transformer-based monocular NOCS prediction model (NOCSformer) that can predict accurate NOCS, instance masks and poses from 2D object detections across diverse classes. It is the first NOCS model that can generalize to a broad range of classes when prompted with 2D boxes. We evaluate our model on the task of 3D oriented bounding box prediction, where it achieves comparable results to state-of-the-art 3D detection methods such as Cube R-CNN. Unlike other 3D detection methods, our model also provides detailed and accurate 3D object shape and segmentation. We propose a novel benchmark for the task of NOCS prediction based on OmniNOCS, which we hope will serve as a useful baseline for future work in this area. Our dataset and code will be at the project website: https://omninocs.github.io., Comment: Accepted to ECCV 2024, project website: https://omninocs.github.io

Enterprise chatbots, powered by generative AI, are emerging as key applications to enhance employee productivity. Retrieval Augmented Generation (RAG), Large Language Models (LLMs), and orchestration frameworks like Langchain and Llamaindex are crucial for building these chatbots. However, creating effective enterprise chatbots is challenging and requires meticulous RAG pipeline engineering. This includes fine-tuning embeddings and LLMs, extracting documents from vector databases, rephrasing queries, reranking results, designing prompts, honoring document access controls, providing concise responses, including references, safeguarding personal information, and building orchestration agents. We present a framework for building RAG-based chatbots based on our experience with three NVIDIA chatbots: for IT/HR benefits, financial earnings, and general content. Our contributions are three-fold: introducing the FACTS framework (Freshness, Architectures, Cost, Testing, Security), presenting fifteen RAG pipeline control points, and providing empirical results on accuracy-latency tradeoffs between large and small LLMs. To the best of our knowledge, this is the first paper of its kind that provides a holistic view of the factors as well as solutions for building secure enterprise-grade chatbots.", Comment: 8 pages, 6 figures, 2 tables, Preprint submission to ACM CIKM 2024

Knowing the effect of an intervention is critical for human decision-making, but current approaches for causal effect estimation rely on manual data collection and structuring, regardless of the causal assumptions. This increases both the cost and time-to-completion for studies. We show how large, diverse observational text data can be mined with large language models (LLMs) to produce inexpensive causal effect estimates under appropriate causal assumptions. We introduce NATURAL, a novel family of causal effect estimators built with LLMs that operate over datasets of unstructured text. Our estimators use LLM conditional distributions (over variables of interest, given the text data) to assist in the computation of classical estimators of causal effect. We overcome a number of technical challenges to realize this idea, such as automating data curation and using LLMs to impute missing information. We prepare six (two synthetic and four real) observational datasets, paired with corresponding ground truth in the form of randomized trials, which we used to systematically evaluate each step of our pipeline. NATURAL estimators demonstrate remarkable performance, yielding causal effect estimates that fall within 3 percentage points of their ground truth counterparts, including on real-world Phase 3/4 clinical trials. Our results suggest that unstructured text data is a rich source of causal effect information, and NATURAL is a first step towards an automated pipeline to tap this resource., Comment: 26 pages, 10 figures

A widespread adoption of Virtual, Augmented, and Mixed Reality (VR/AR/MR), collectively referred to as Extended Reality (XR), has become a tangible possibility to revolutionize educational and training scenarios by offering immersive, interactive experiences. In this paper we present \textsf{INTERACT}, an authoring tool for creating advanced 3D physics-based Intelligent Tutoring Systems (ITS) by individual developers or small-scale development teams. \textsf{INTERACT} is based on a cutting edge physics engine allowing realistic interactions such as collision detection and ergonomic evaluations. We demonstrate the benefits of \textsf{INTERACT} by developing a set of training scenarios for a use case of a Laser cutting machine. The use case illustrates the numerous possibilities such as creating interaction with objects, ease of configuring a scenario and how to design the visual effects to the machine.

Bindings for machine learning frameworks (such as TensorFlow and PyTorch) allow developers to integrate a framework's functionality using a programming language different from the framework's default language (usually Python). In this paper, we study the impact of using TensorFlow and PyTorch bindings in C#, Rust, Python and JavaScript on the software quality in terms of correctness (training and test accuracy) and time cost (training and inference time) when training and performing inference on five widely used deep learning models. Our experiments show that a model can be trained in one binding and used for inference in another binding for the same framework without losing accuracy. Our study is the first to show that using a non-default binding can help improve machine learning software quality from the time cost perspective compared to the default Python binding while still achieving the same level of correctness.

Automatically locating a bug within a large codebase remains a significant challenge for developers. Existing techniques often struggle with generalizability and deployment due to their reliance on application-specific data and large model sizes. This paper proposes a novel pre-trained language model (PLM) based technique for bug localization that transcends project and language boundaries. Our approach leverages contrastive learning to enhance the representation of bug reports and source code. It then utilizes a novel ranking approach that combines commit messages and code segments. Additionally, we introduce a knowledge distillation technique that reduces model size for practical deployment without compromising performance. This paper presents several key benefits. By incorporating code segment and commit message analysis alongside traditional file-level examination, our technique achieves better bug localization accuracy. Furthermore, our model excels at generalizability - trained on code from various projects and languages, it can effectively identify bugs in unseen codebases. To address computational limitations, we propose a CPU-compatible solution. In essence, proposed work presents a highly effective, generalizable, and efficient bug localization technique with the potential to real-world deployment.

We report on the results of an image-based search for pulsar candidates toward the Galactic bulge. We used mosaic images from the MeerKAT radio telescope, that were taken as part of a 173 deg**2 survey of the bulge and Galactic center of our Galaxy at L band (856-1712 MHz) in all four Stokes I, Q, U and V. The image root-mean-square noise levels of 12-17 uJy/ba represent a significant increase in sensitivity over past image-based pulsar searches. Our primary search criterion was circular polarization, but we used other criteria including linear polarization, in-band spectral index, compactness, variability and multi-wavelength counterparts to select pulsar candidates. We first demonstrate the efficacy of this technique by searching for polarized emission from known pulsars, and comparing our results with measurements from the literature. Our search resulted in a sample of 75 polarized pulsar candidates. Bright stars or young stellar objects were associated with 28 of these sources, including a small sample of highly polarized dwarf stars with pulsar-like steep spectra. Comparing the properties of this sample with the known pulsars, we identified 30 compelling candidates for pulsation follow-up, including two sources with both strong circular and linear polarization. The remaining 17 sources are either pulsars or stars, but we cannot rule out an extragalactic origin or image artifacts among the brighter, flat spectrum objects., Comment: 28 pages, 10 figures, 6 tables. ApJ, submitted

Data clones are defined as multiple copies of the same data among datasets. Presence of data clones between datasets can cause issues such as difficulties in managing data assets and data license violations when using datasets with clones to build AI software. However, detecting data clones is not trivial. Majority of the prior studies in this area rely on structural information to detect data clones (e.g., font size, column header). However, tabular datasets used to build AI software are typically stored without any structural information. In this paper, we propose a novel method called SimClone for data clone detection in tabular datasets without relying on structural information. SimClone method utilizes value similarities for data clone detection. We also propose a visualization approach as a part of our SimClone method to help locate the exact position of the cloned data between a dataset pair. Our results show that our SimClone outperforms the current state-of-the-art method by at least 20\% in terms of both F1-score and AUC. In addition, SimClone's visualization component helps identify the exact location of the data clone in a dataset with a Precision@10 value of 0.80 in the top 20 true positive predictions., Comment: 24 pages, 9 figures

In this project, we investigate the accuracy of forecasting intraday and daily trading volume of the exchange-traded fund SPY. The ability to forecast volume over varying time intervals with high accuracy is a critical element to many trading strategies. After performing exploratory data analysis on intraday and daily SPY data we identify three methods for our analysis: ARIMA and ARIMAX models, with or without seasonality, as well as a Frequency Domain Process Representation. To evaluate predictive power of our models, we use mean squared error, mean absolute percentage error, and volume weighted average price (VWAP) tracking error. All models for both intraday and daily data output strong VWAP predictions in comparison to the VWAP estimates produced by naive baseline methodologies. In both cases volume is most accurately forecasted using ARIMA models with exogenous variables in the form of technical indicators, with intraday incorporating a seasonal component and daily not.

The spherical mean transform associates to a function $f$ its integral averages over all spheres. We consider the spherical mean transform for functions supported in the unit ball $\mathbb{B}$ in $\mathbb{R}^n$ for odd $n$, with the centers of integration spheres restricted to the unit sphere $\mathbb{S}^{n-1}$. In this setup, Rubin employed properties of Erd\'elyi-Kober fractional integrals and analytic continuation to re-derive the explicit inversion formulas proved earlier by Finch, Patch, and Rakesh using wave equation techniques. As part of his work, Rubin stated a conjecture relating spherical mean transform, its associated backprojection operator and the Riesz potential. Furthermore, he pointed to the necessity of a detailed analysis of injectivity of the backprojection operator as a crucial step toward the resolution of his conjecture. This article addresses both questions posed by Rubin by providing a characterization of the null space of the backprojection operator, and disproving the conjecture through the construction of an explicit counterexample. Crucial to the proofs is the range characterization for the spherical mean transform in odd dimensions derived recently by the authors.

We consider a wireless distributed computing system based on the MapReduce framework, which consists of three phases: \textit{Map}, \textit{Shuffle}, and \textit{Reduce}. The system consists of a set of distributed nodes assigned to compute arbitrary output functions depending on a file library. The computation of the output functions is decomposed into Map and Reduce functions, and the Shuffle phase, which involves the data exchange, links the two. In our model, the Shuffle phase communication happens over a full-duplex wireless interference channel. For this setting, a coded wireless MapReduce distributed computing scheme exists in the literature, achieving optimal performance under one-shot linear schemes. However, the scheme requires the number of input files to be very large, growing exponentially with the number of nodes. We present schemes that require the number of files to be in the order of the number of nodes and achieve the same performance as the existing scheme. The schemes are obtained by designing a structure called wireless MapReduce array that succinctly represents all three phases in a single array. The wireless MapReduce arrays can also be obtained from the extended placement delivery arrays known for multi-antenna coded caching schemes., Comment: Communicated to 2024 IEEE Information Theory Workshop (ITW'24), Shenzhen, China

Utilizing air-traffic control (ATC) data for downstream natural-language processing tasks requires preprocessing steps. Key steps are the transcription of the data via automatic speech recognition (ASR) and speaker diarization, respectively speaker role detection (SRD) to divide the transcripts into pilot and air-traffic controller (ATCO) transcripts. While traditional approaches take on these tasks separately, we propose a transformer-based joint ASR-SRD system that solves both tasks jointly while relying on a standard ASR architecture. We compare this joint system against two cascaded approaches for ASR and SRD on multiple ATC datasets. Our study shows in which cases our joint system can outperform the two traditional approaches and in which cases the other architectures are preferable. We additionally evaluate how acoustic and lexical differences influence all architectures and show how to overcome them for our joint architecture., Comment: Accepted at Interspeech 2024

We prove that the disordered monomer-dimer model does not admit infinite volume incongruent ground states in $\mathbb{Z}^d$ which can be obtained as a limit of finite volume ground states. Furthermore, we also prove that these ground states are stable under perturbation of the weights in a precise sense. As an application, we obtain a CLT for the ground state weight for a growing sequence of tori. Our motivation stems from a similar and long standing open question for the short range Edwards-Anderson spin glass model.

While existing literature relies on performance differences to uncover gender biases in ASR models, a deeper analysis is essential to understand how gender is encoded and utilized during transcript generation. This work investigates the encoding and utilization of gender in the latent representations of two transformer-based ASR models, Wav2Vec2 and HuBERT. Using linear erasure, we demonstrate the feasibility of removing gender information from each layer of an ASR model and show that such an intervention has minimal impacts on the ASR performance. Additionally, our analysis reveals a concentration of gender information within the first and last frames in the final layers, explaining the ease of erasing gender in these layers. Our findings suggest the prospect of creating gender-neutral embeddings that can be integrated into ASR frameworks without compromising their efficacy., Comment: Accepted at Interspeech 2024

Generic Face Image Quality Assessment (GFIQA) evaluates the perceptual quality of facial images, which is crucial in improving image restoration algorithms and selecting high-quality face images for downstream tasks. We present a novel transformer-based method for GFIQA, which is aided by two unique mechanisms. First, a Dual-Set Degradation Representation Learning (DSL) mechanism uses facial images with both synthetic and real degradations to decouple degradation from content, ensuring generalizability to real-world scenarios. This self-supervised method learns degradation features on a global scale, providing a robust alternative to conventional methods that use local patch information in degradation learning. Second, our transformer leverages facial landmarks to emphasize visually salient parts of a face image in evaluating its perceptual quality. We also introduce a balanced and diverse Comprehensive Generic Face IQA (CGFIQA-40k) dataset of 40K images carefully designed to overcome the biases, in particular the imbalances in skin tone and gender representation, in existing datasets. Extensive analysis and evaluation demonstrate the robustness of our method, marking a significant improvement over prior methods., Comment: Accepted by CVPR 2024, Project Page: https://dsl-fiqa.github.io/

Entity matching is the task of linking records from different sources that refer to the same real-world entity. Past work has primarily treated entity linking as a standard supervised learning problem. However, supervised entity matching models often do not generalize well to new data, and collecting exhaustive labeled training data is often cost prohibitive. Further, recent efforts have adopted LLMs for this task in few/zero-shot settings, exploiting their general knowledge. But LLMs are prohibitively expensive for performing inference at scale for real-world entity matching tasks. As an efficient alternative, we re-cast entity matching as a conditional generation task as opposed to binary classification. This enables us to "distill" LLM reasoning into smaller entity matching models via natural language explanations. This approach achieves strong performance, especially on out-of-domain generalization tests (10.85% F-1) where standalone generative methods struggle. We perform ablations that highlight the importance of explanations, both for performance and model robustness.

Foundational vision transformer models have shown impressive few shot performance on many vision tasks. This research presents a novel investigation into the application of parameter efficient fine-tuning methods within an active learning (AL) framework, to advance the sampling selection process in extremely budget constrained classification tasks. The focus on image datasets, known for their out-of-distribution characteristics, adds a layer of complexity and relevance to our study. Through a detailed evaluation, we illustrate the improved AL performance on these challenging datasets, highlighting the strategic advantage of merging parameter efficient fine tuning methods with foundation models. This contributes to the broader discourse on optimizing AL strategies, presenting a promising avenue for future exploration in leveraging foundation models for efficient and effective data annotation in specialized domains., Comment: Accepted for CVPR2024 Transformers for Vision Workshop

The development of fault-tolerant quantum computers (FTQCs) is gaining increased attention within the quantum computing community. Like their digital counterparts, FTQCs, equipped with error correction and large qubit numbers, promise to solve some of humanity's grand challenges. Estimates of the resource requirements for future FTQC systems are essential to making design choices and prioritizing R&D efforts to develop critical technologies. Here, we present a resource estimation framework and software tool that estimates the physical resources required to execute specific quantum algorithms, compiled into their graph-state form, and laid out onto a modular superconducting hardware architecture. This tool can predict the size, power consumption, and execution time of these algorithms at as they approach utility-scale according to explicit assumptions about the system's physical layout, thermal load, and modular connectivity. We use this tool to study the total resources on a proposed modular architecture and the impact of tradeoffs between and inter-module connectivity, latency and resource requirements.

Aggression is a two-player game of troop placement and attack played on a map (modeled as a graph). Players take turns deploying troops on a territory (a vertex on the graph) until they run out. Once all troops are placed, players take turns attacking enemy territories. A territory can be attacked if it has $k$ troops and there are more than $k$ enemy troops on adjacent territories. At the end of the game, the player who controls the most territories wins. In the case of a tie, the player with more surviving troops wins. The first player to exhaust their troops in the placement phase leads the attack phase. We study the complexity of the game when the graph along with an assignment of troops and the sequence of attacks planned by the second player. Even in this restrained setting, we show that the problem of determining an optimal sequence of first player moves is NP-complete. We then analyze the game for when the input graph is a matching or a cycle., Comment: 19 pages, 4 figures; a shorter version was accepted for presentation at COCOON 2025

This paper forms the second of a two-part series on the value of a participatory approach to AI development and deployment. The first paper had crafted a principled, as well as pragmatic, justification for deploying participatory methods in these two exercises (that is, development and deployment of AI). The pragmatic justification is that it improves the quality of the overall algorithm by providing more granular and minute information. The more principled justification is that it offers a voice to those who are going to be affected by the deployment of the algorithm, and through engagement attempts to build trust and buy-in for an AI system. By a participatory approach, we mean including various stakeholders (defined a certain way) in the actual decision making process through the life cycle of an AI system. Despite the justifications offered above, actual implementation depends crucially on how stakeholders in the entire process are identified, what information is elicited from them, and how it is incorporated. This paper will test these preliminary conclusions in two sectors, the use of facial recognition technology in the upkeep of law and order and the use of large language models in the healthcare sector. These sectors have been chosen for two primary reasons. Since Facial Recognition Technologies are a branch of AI solutions that are well-researched and the impact of which is well documented, it provides an established space to illustrate the various aspects of adapting PAI to an existing domain, especially one that has been quite contentious in the recent past. LLMs in healthcare provide a canvas for a relatively less explored space, and helps us illustrate how one could possibly envision enshrining the principles of PAI for a relatively new technology, in a space where innovation must always align with patient welfare.

The widespread adoption of Artificial Intelligence (AI) technologies in the public and private sectors has resulted in them significantly impacting the lives of people in new and unexpected ways. In this context, it becomes important to inquire how their design, development and deployment takes place. Upon this inquiry, it is seen that persons who will be impacted by the deployment of these systems have little to no say in how they are developed. Seeing this as a lacuna, this research study advances the premise that a participatory approach is beneficial (both practically and normatively) to building and using more responsible, safe, and human-centric AI systems. Normatively, it enhances the fairness of the process and empowers citizens in voicing concerns to systems that may heavily impact their lives. Practically, it provides developers with new avenues of information which will be beneficial to them in improving the quality of the AI algorithm. The paper advances this argument first, by describing the life cycle of an AI system; second, by identifying criteria which may be used to identify relevant stakeholders for a participatory exercise; and third, by mapping relevant stakeholders to different stages of AI lifecycle. This paper forms the first part of a two-part series on participatory governance in AI. The second paper will expand upon and concretise the principles developed in this paper and apply the same to actual use cases of AI systems.

While diffusion models have significantly advanced the quality of image generation their capability to accurately and coherently render text within these images remains a substantial challenge. Conventional diffusion-based methods for scene text generation are typically limited by their reliance on an intermediate layout output. This dependency often results in a constrained diversity of text styles and fonts an inherent limitation stemming from the deterministic nature of the layout generation phase. To address these challenges this paper introduces SceneTextGen a novel diffusion-based model specifically designed to circumvent the need for a predefined layout stage. By doing so SceneTextGen facilitates a more natural and varied representation of text. The novelty of SceneTextGen lies in its integration of three key components: a character-level encoder for capturing detailed typographic properties coupled with a character-level instance segmentation model and a word-level spotting model to address the issues of unwanted text generation and minor character inaccuracies. We validate the performance of our method by demonstrating improved character recognition rates on generated images across different public visual text datasets in comparison to both standard diffusion based methods and text specific methods., Comment: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2024, pp. 7496-7506

3D generative models have shown significant promise in structure-based drug design (SBDD), particularly in discovering ligands tailored to specific target binding sites. Existing algorithms often focus primarily on ligand-target binding, characterized by binding affinity. Moreover, models trained solely on target-ligand distribution may fall short in addressing the broader objectives of drug discovery, such as the development of novel ligands with desired properties like drug-likeness, and synthesizability, underscoring the multifaceted nature of the drug design process. To overcome these challenges, we decouple the problem into molecular generation and property prediction. The latter synergistically guides the diffusion sampling process, facilitating guided diffusion and resulting in the creation of meaningful molecules with the desired properties. We call this guided molecular generation process as TAGMol. Through experiments on benchmark datasets, TAGMol demonstrates superior performance compared to state-of-the-art baselines, achieving a 22% improvement in average Vina Score and yielding favorable outcomes in essential auxiliary properties. This establishes TAGMol as a comprehensive framework for drug generation.

Tabular data are omnipresent in various sectors of industries. Neural networks for tabular data such as TabNet have been proposed to make predictions while leveraging the attention mechanism for interpretability. However, the inferred attention masks are often dense, making it challenging to come up with rationales about the predictive signal. To remedy this, we propose InterpreTabNet, a variant of the TabNet model that models the attention mechanism as a latent variable sampled from a Gumbel-Softmax distribution. This enables us to regularize the model to learn distinct concepts in the attention masks via a KL Divergence regularizer. It prevents overlapping feature selection by promoting sparsity which maximizes the model's efficacy and improves interpretability to determine the important features when predicting the outcome. To assist in the interpretation of feature interdependencies from our model, we employ a large language model (GPT-4) and use prompt engineering to map from the learned feature mask onto natural language text describing the learned signal. Through comprehensive experiments on real-world datasets, we demonstrate that InterpreTabNet outperforms previous methods for interpreting tabular data while attaining competitive accuracy., Comment: ICML 2024 Spotlight

Neural operators extend data-driven models to map between infinite-dimensional functional spaces. While these operators perform effectively in either the time or frequency domain, their performance may be limited when applied to non-stationary spatial or temporal signals whose frequency characteristics change with time. Here, we introduce Complex Neural Operator (CoNO) that parameterizes the integral kernel using Fractional Fourier Transform (FrFT), better representing non-stationary signals in a complex-valued domain. Theoretically, we prove the universal approximation capability of CoNO. We perform an extensive empirical evaluation of CoNO on seven challenging partial differential equations (PDEs), including regular grids, structured meshes, and point clouds. Empirically, CoNO consistently attains state-of-the-art performance, showcasing an average relative gain of 10.9%. Further, CoNO exhibits superior performance, outperforming all other models in additional tasks such as zero-shot super-resolution and robustness to noise. CoNO also exhibits the ability to learn from small amounts of data -- giving the same performance as the next best model with just 60% of the training data. Altogether, CoNO presents a robust and superior model for modeling continuous dynamical systems, providing a fillip to scientific machine learning., Comment: Under Review

The three-body dissociation dynamics of the dicationic camphor molecule (C$_{10}$H$_{16}$O$^{2+}$) resulting from Auger decay are investigated using soft X-ray synchrotron radiation. A photoelectron-photoion-photoion coincidence (PEPIPICO) method, a combination of a velocity map imaging (VMI) spectrometer and a time-of-flight (ToF) spectrometer is employed to measure the 3D momenta of ions detected in coincidence. The ion mass spectra and the ion-ion coincidence map at photon energies of 287.9 eV (below the C 1s ionization potential) and 292.4 eV (above the C 1s ionization potential for skeletal carbon) reveal that fragmentation depends on the final dicationic state rather than the initial excitation. Using the native frame method, three new fragmentation channels are discussed; (1) CH$_2$CO$^+$ + C$_7$H$_{11}^+$ + CH$_3$, (2) CH$_3^+$ + C$_7$H$_{11}^+$ + CH$_2$CO, and (3) C$_2$H$_5^+$ + C$_6$H$_9^+$ + CH$_2$CO. The dominating nature of sequential decay with deferred charge separation is clearly evidenced in all three channels. The results are discussed based on the experimental angular distributions and momenta distributions, corroborated by geometry optimization of the ground, monocationic, and dicationic camphor molecule.

It is important for researchers to understand precisely how data scientists turn raw data into insights, including typical programming patterns, workflow, and methodology. This paper contributes a novel system, called DataInquirer, that tracks incremental code executions in Jupyter notebooks (a type of computational notebook). The system allows us to quantitatively measure timing, workflow, and operation frequency in data science tasks without resorting to human annotation or interview. In a series of pilot studies, we collect 97 traces, logging data scientist activities across four studies. While this paper presents a general system and data analysis approach, we focus on a foundational sub-question in our pilot studies: How consistent are different data scientists in analyzing the same data? We taxonomize variation between data scientists on the same dataset according to three categories: semantic, syntactic, and methodological. Our results suggest that there are statistically significant differences in the conclusions reached by different data scientists on the same task and present quantitative evidence for this phenomenon. Furthermore, our results suggest that AI-powered code tools subtly influence these results, allowing student participants to generate workflows that more resemble expert data practitioners.

Model-Free Reinforcement Learning (MFRL), leveraging the policy gradient theorem, has demonstrated considerable success in continuous control tasks. However, these approaches are plagued by high gradient variance due to zeroth-order gradient estimation, resulting in suboptimal policies. Conversely, First-Order Model-Based Reinforcement Learning (FO-MBRL) methods employing differentiable simulation provide gradients with reduced variance but are susceptible to sampling error in scenarios involving stiff dynamics, such as physical contact. This paper investigates the source of this error and introduces Adaptive Horizon Actor-Critic (AHAC), an FO-MBRL algorithm that reduces gradient error by adapting the model-based horizon to avoid stiff dynamics. Empirical findings reveal that AHAC outperforms MFRL baselines, attaining 40% more reward across a set of locomotion tasks and efficiently scaling to high-dimensional control environments with improved wall-clock-time efficiency., Comment: Website https://adaptive-horizon-actor-critic.github.io/

Tracking data lineage is important for data integrity, reproducibility, and debugging data science workflows. However, fine-grained lineage (i.e., at a cell level) is challenging to store, even for the smallest datasets. This paper introduces DSLog, a storage system that efficiently stores, indexes, and queries array data lineage, agnostic to capture methodology. A main contribution is our new compression algorithm, named ProvRC, that compresses captured lineage relationships. Using ProvRC for lineage compression result in a significant storage reduction over functions with simple spatial regularity, beating alternative columnar-store baselines by up to 2000x}. We also show that ProvRC facilitates in-situ query processing that allows forward and backward lineage queries without decompression - in the optimal case, surpassing baselines by 20x in query latency on random numpy pipelines.

With the goal of identifying common practices in data science projects, this paper proposes a framework for logging and understanding incremental code executions in Jupyter notebooks. This framework aims to allow reasoning about how insights are generated in data science and extract key observations into best data science practices in the wild. In this paper, we show an early prototype of this framework and ran an experiment to log a machine learning project for 25 undergraduate students.

Video analytics systems based on deep learning models are often opaque and brittle and require explanation systems to help users debug. Current model explanation system are very good at giving literal explanations of behavior in terms of pixel contributions but cannot integrate information about the physical or systems processes that might influence a prediction. This paper introduces the idea that a simple form of causal reasoning, called a regression discontinuity design, can be used to associate changes in multiple key performance indicators to physical real world phenomena to give users a more actionable set of video analytics explanations. We overview the system architecture and describe a vision of the impact that such a system might have.

In an era where digital platforms increasingly mediate public discourse, grasping the complexities and nuances in affective polarization--especially as influenced by key figures on social media--has never been more vital. This study delves into the intricate web of interactions on Twitter, now rebranded as 'X', to unravel how influencer-led conversations catalyze shifts in public sentiment, laying bare the complex dynamics that underpin online polarization. Employing a novel methodological framework that includes counterfactual analysis, we analyze scenarios with and without specific influencer-led conversations. Our findings illuminate the significant role influencers play in shaping public discourse, offering insights into the mechanisms of online polarization and suggesting pathways for future research to mitigate divisiveness and explore new methods for quantifying affective polarization. This research contributes to the broader understanding of digital communication's impact on societal polarization, underscoring the importance of detailed analysis in developing strategies to foster a more cohesive digital public sphere., Comment: 8 pages, 4 figures

Embeddings produced by pre-trained deep neural networks (DNNs) are widely used; however, their efficacy for downstream tasks can vary widely. We study the factors influencing out-of-distribution (OOD) generalization of pre-trained DNN embeddings through the lens of the tunnel effect hypothesis, which suggests deeper DNN layers compress representations and hinder OOD performance. Contrary to earlier work, we find the tunnel effect is not universal. Based on 10,584 linear probes, we study the conditions that mitigate the tunnel effect by varying DNN architecture, training dataset, image resolution, and augmentations. We quantify each variable's impact using a novel SHAP analysis. Our results emphasize the danger of generalizing findings from toy datasets to broader contexts., Comment: Project Website: https://yousuf907.github.io/oodg

Transformers have significantly advanced the modeling of Electronic Health Records (EHR), yet their deployment in real-world healthcare is limited by several key challenges. Firstly, the quadratic computational cost and insufficient context length of these models pose significant obstacles for hospitals in processing the extensive medical histories typical in EHR data. Additionally, existing models employ separate finetuning for each clinical task, complicating maintenance in healthcare environments. Moreover, these models focus exclusively on either clinical prediction or EHR forecasting, lacking the flexibility to perform well across both. To overcome these limitations, we introduce EHRMamba, a robust foundation model built on the Mamba architecture. EHRMamba can process sequences up to four times longer than previous models due to its linear computational cost. We also introduce a novel approach to Multitask Prompted Finetuning (MTF) for EHR data, which enables EHRMamba to simultaneously learn multiple clinical tasks in a single finetuning phase, significantly enhancing deployment and cross-task generalization. Furthermore, our model leverages the HL7 FHIR data standard to simplify integration into existing hospital systems. Alongside EHRMamba, we open-source Odyssey, a toolkit designed to support the development and deployment of EHR foundation models, with an emphasis on data standardization and interpretability. Our evaluations on the MIMIC-IV dataset demonstrate that EHRMamba advances state-of-the-art performance across 6 major clinical tasks and excels in EHR forecasting, marking a significant leap forward in the field., Comment: 17 Pages, 4 Figures