Search

The Solar Ultraviolet Imaging Telescope (SUIT) is an instrument on the Aditya-L1 mission of the Indian Space Research Organization (ISRO) launched on September 02, 2023. SUIT continuously provides, near-simultaneous full-disk and region-of-interest images of the Sun, slicing through the photosphere and chromosphere and covering a field of view up to 1.5 solar radii. For this purpose, SUIT uses 11 filters tuned at different wavelengths in the 200{--}400~nm range, including the Mg~{\sc ii} h~and~k and Ca~{\sc ii}~H spectral lines. The observations made by SUIT help us understand the magnetic coupling of the lower and middle solar atmosphere. In addition, for the first time, it allows the measurements of spatially resolved solar broad-band radiation in the near and mid ultraviolet, which will help constrain the variability of the solar ultraviolet irradiance in a wavelength range that is central for the chemistry of the Earth's atmosphere. This paper discusses the details of the instrument and data products., Comment: 37 pages, Accepted for Publication in Solar Physics

Preserving topological structures is important in real-world applications, particularly in sensitive domains such as healthcare and medicine, where the correctness of human anatomy is critical. However, most existing image editing models focus on manipulating intensity and texture features, often overlooking object geometry within images. To address this issue, this paper introduces a novel method, Topology-Preserved Image Editing with text instructions (TPIE), that for the first time ensures the topology and geometry remaining intact in edited images through text-guided generative diffusion models. More specifically, our method treats newly generated samples as deformable variations of a given input template, allowing for controllable and structure-preserving edits. Our proposed TPIE framework consists of two key modules: (i) an autoencoder-based registration network that learns latent representations of object transformations, parameterized by velocity fields, from pairwise training images; and (ii) a novel latent conditional geometric diffusion (LCDG) model efficiently capturing the data distribution of learned transformation features conditioned on custom-defined text instructions. We validate TPIE on a diverse set of 2D and 3D images and compare them with state-of-the-art image editing approaches. Experimental results show that our method outperforms other baselines in generating more realistic images with well-preserved topology. Our code will be made publicly available on Github.

Physically motivated stochastic dynamics are often used to sample from high-dimensional distributions. However such dynamics often get stuck in specific regions of their state space and mix very slowly to the desired stationary state. This causes such systems to approximately sample from a metastable distribution which is usually quite different from the desired, stationary distribution of the dynamic. We rigorously show that, in the case of multi-variable discrete distributions, the true model describing the stationary distribution can be recovered from samples produced from a metastable distribution under minimal assumptions about the system. This follows from a fundamental observation that the single-variable conditionals of metastable distributions that satisfy a strong metastability condition are on average close to those of the stationary distribution. This holds even when the metastable distribution differs considerably from the true model in terms of global metrics like Kullback-Leibler divergence or total variation distance. This property allows us to learn the true model using a conditional likelihood based estimator, even when the samples come from a metastable distribution concentrated in a small region of the state space. Explicit examples of such metastable states can be constructed from regions that effectively bottleneck the probability flow and cause poor mixing of the Markov chain. For specific cases of binary pairwise undirected graphical models (i.e. Ising models), we extend our results to further rigorously show that data coming from metastable states can be used to learn the parameters of the energy function and recover the structure of the model., Comment: Updated version, 31 pages

QR codes have become ubiquitous in daily life, enabling rapid information exchange. With the increasing adoption of smart wearable devices, there is a need for efficient, and friction-less QR code reading capabilities from Egocentric point-of-views. However, adapting existing phone-based QR code readers to egocentric images poses significant challenges. Code reading from egocentric images bring unique challenges such as wide field-of-view, code distortion and lack of visual feedback as compared to phones where users can adjust the position and framing. Furthermore, wearable devices impose constraints on resources like compute, power and memory. To address these challenges, we present EgoQR, a novel system for reading QR codes from egocentric images, and is well suited for deployment on wearable devices. Our approach consists of two primary components: detection and decoding, designed to operate on high-resolution images on the device with minimal power consumption and added latency. The detection component efficiently locates potential QR codes within the image, while our enhanced decoding component extracts and interprets the encoded information. We incorporate innovative techniques to handle the specific challenges of egocentric imagery, such as varying perspectives, wider field of view, and motion blur. We evaluate our approach on a dataset of egocentric images, demonstrating 34% improvement in reading the code compared to an existing state of the art QR code readers., Comment: Submitted to ICLR 2025

Autonomous driving at unsignalized intersections is still considered a challenging application for machine learning due to the complications associated with handling complex multi-agent scenarios characterized by a high degree of uncertainty. Automating the decision-making process at these safety-critical environments involves comprehending multiple levels of abstractions associated with learning robust driving behaviors to enable the vehicle to navigate efficiently. In this survey, we aim at exploring the state-of-the-art techniques implemented for decision-making applications, with a focus on algorithms that combine Reinforcement Learning (RL) and deep learning for learning traversing policies at unsignalized intersections. The reviewed schemes vary in the proposed driving scenario, in the assumptions made for the used intersection model, in the tackled challenges, and in the learning algorithms that are used. We have presented comparisons for these techniques to highlight their limitations and strengths. Based on our in-depth investigation, it can be discerned that a robust decision-making scheme for navigating real-world unsignalized intersection has yet to be developed. Along with our analysis and discussion, we recommend potential research directions encouraging the interested players to tackle the highlighted challenges. By adhering to our recommendations, decision-making architectures that are both non-overcautious and safe, yet feasible, can be trained and validated in real-world unsignalized intersections environments.

Let $\mathscr{B}_n = \{ \pm x_1, \pm x_2, \pm x_3, \cdots, \pm x_{n-1}, x_n \}$ where $n>1$ is fixed, $x_i \in \mathbb{R}^+$, $i = 1, 2, 3, \cdots, n$ and $x_1 < x_2 < x_3 < \cdots < x_n$. Let $\phi(\mathscr{B}_n)$ be the set of all non-empty subsets $S = \{u_1, u_2,\cdots, u_t\}$ of $\mathscr{B}_n$ such that $|u_1|<|u_2|<\cdots <|u_{t-1}|

Fairness in AI and ML systems is increasingly linked to the proper treatment and recognition of data workers involved in training dataset development. Yet, those who collect and annotate the data, and thus have the most intimate knowledge of its development, are often excluded from critical discussions. This exclusion prevents data annotators, who are domain experts, from contributing effectively to dataset contextualization. Our investigation into the hiring and engagement practices of 52 data work requesters on platforms like Amazon Mechanical Turk reveals a gap: requesters frequently hold naive or unchallenged notions of worker identities and capabilities and rely on ad-hoc qualification tasks that fail to respect the workers' expertise. These practices not only undermine the quality of data but also the ethical standards of AI development. To rectify these issues, we advocate for policy changes to enhance how data annotation tasks are designed and managed and to ensure data workers are treated with the respect they deserve., Comment: Accepted for publication at AIES 2024

We present a comprehensive theoretical framework governing the dynamics of droplets in acoustic fields, applicable to all droplet sizes, from the Rayleigh limit (D<

Although quantum computing holds promise to accelerate a wide range of computational tasks, the quantum simulation of quantum dynamics as originally envisaged by Feynman remains the most promising candidate for achieving quantum advantage. A less explored possibility with comparably far-reaching technological applicability is the quantum simulation of classical nonlinear dynamics. Attempts to develop digital quantum algorithms based on the Koopman von Neumann formalism have met with challenges because of the necessary projection step from an infinite-dimensional Hilbert space to the finite-dimensional subspace described by a collection of qubits. This finitization produces numerical artifacts that limit solutions to very short time horizons. In this paper we review continuous-variable quantum computing (CVQC), which naturally avoids such obstacles, and a CVQC algorithm for KvN simulation of classical nonlinear dynamics is advocated. In particular, we present explicit gate synthesis for product-formula Hamiltonian simulation of anharmonic vibrational dynamics., Comment: 11 pages

Motion and deformation analysis of cardiac magnetic resonance (CMR) imaging videos is crucial for assessing myocardial strain of patients with abnormal heart functions. Recent advances in deep learning-based image registration algorithms have shown promising results in predicting motion fields from routinely acquired CMR sequences. However, their accuracy often diminishes in regions with subtle appearance changes, with errors propagating over time. Advanced imaging techniques, such as displacement encoding with stimulated echoes (DENSE) CMR, offer highly accurate and reproducible motion data but require additional image acquisition, which poses challenges in busy clinical flows. In this paper, we introduce a novel Latent Motion Diffusion model (LaMoD) to predict highly accurate DENSE motions from standard CMR videos. More specifically, our method first employs an encoder from a pre-trained registration network that learns latent motion features (also considered as deformation-based shape features) from image sequences. Supervised by the ground-truth motion provided by DENSE, LaMoD then leverages a probabilistic latent diffusion model to reconstruct accurate motion from these extracted features. Experimental results demonstrate that our proposed method, LaMoD, significantly improves the accuracy of motion analysis in standard CMR images; hence improving myocardial strain analysis in clinical settings for cardiac patients. Our code is publicly available at https://github.com/jr-xing/LaMoD.

The PROSPECT experiment is designed to perform precise searches for antineutrino disappearance at short distances (7 - 9~m) from compact nuclear reactor cores. This Letter reports results from a new neutrino oscillation analysis performed using the complete data sample from the PROSPECT-I detector operated at the High Flux Isotope Reactor in 2018. The analysis uses a multi-period selection of inverse beta decay neutrino interactions with reduced backgrounds and enhanced statistical power to set limits on electron-flavor disappearance caused by mixing with sterile neutrinos with 0.2 - 20 eV$^2$ mass splittings. Inverse beta decay positron energy spectra from six different reactor-detector distance ranges are found to be statistically consistent with one another, as would be expected in the absence of sterile neutrino oscillations. The data excludes at 95% confidence level the existence of sterile neutrinos in regions above 3~eV$^2$ previously unexplored by terrestrial experiments, including all space below 10~eV$^2$ suggested by the recently strengthened Gallium Anomaly. The best-fit point of the Neutrino-4 reactor experiment's claimed observation of short-baseline oscillation is ruled out at more than five standard deviations., Comment: 6 pages, 4 figures

The PROSPECT-I detector has several features that enable measurement of the direction of a compact neutrino source. In this paper, a detailed report on the directional measurements made on electron antineutrinos emitted from the High Flux Isotope Reactor is presented. With an estimated true neutrino (reactor to detector) direction of $\phi = 40.8\unicode{xB0} \pm 0.7\unicode{xB0}$ and $\theta = 98.6\unicode{xB0} \pm 0.4\unicode{xB0}$, the PROSPECT-I detector is able to reconstruct an average neutrino direction of $\phi = 39.4\unicode{xB0} \pm 2.9\unicode{xB0}$ and $\theta = 97.6\unicode{xB0} \pm 1.6\unicode{xB0}$. This measurement is made with approximately 48000 Inverse Beta Decay signal events and is the most precise directional reconstruction of reactor antineutrinos to date.

Visual Question Answering (VQA) is an important task in multimodal AI, and it is often used to test the ability of vision-language models to understand and reason on knowledge present in both visual and textual data. However, most of the current VQA models use datasets that are primarily focused on English and a few major world languages, with images that are typically Western-centric. While recent efforts have tried to increase the number of languages covered on VQA datasets, they still lack diversity in low-resource languages. More importantly, although these datasets often extend their linguistic range via translation or some other approaches, they usually keep images the same, resulting in narrow cultural representation. To address these limitations, we construct CVQA, a new Culturally-diverse multilingual Visual Question Answering benchmark, designed to cover a rich set of languages and cultures, where we engage native speakers and cultural experts in the data collection process. As a result, CVQA includes culturally-driven images and questions from across 30 countries on four continents, covering 31 languages with 13 scripts, providing a total of 10k questions. We then benchmark several Multimodal Large Language Models (MLLMs) on CVQA, and show that the dataset is challenging for the current state-of-the-art models. This benchmark can serve as a probing evaluation suite for assessing the cultural capability and bias of multimodal models and hopefully encourage more research efforts toward increasing cultural awareness and linguistic diversity in this field., Comment: 38th Conference on Neural Information Processing Systems (NeurIPS 2024) Track on Datasets and Benchmarks

The emergence of quantum computing technology over the last decade indicates the potential for a transformational impact in the study of quantum mechanical systems. It is natural to presume that such computing technologies would be valuable to large scientific institutions, such as United States national laboratories. However, detailed descriptions of what these institutions would like to use these computers for are limited. To help provide some initial insights into this topic, this report develops detailed use cases of how quantum computing technology could be utilized to enhance a variety of quantum physics research activities at Los Alamos National Laboratory, including quantum magnetic materials, high-temperature superconductivity and nuclear astrophysics simulations. The report discusses how current high-performance computers are used for scientific discovery today and develops detailed descriptions of the types of quantum physics simulations that Los Alamos National Laboratory scientists would like to conduct, if a sufficient computing technology became available. While the report strives to highlight the breadth of potential application areas for quantum computation, this investigation has also indicated that many more use cases exist at Los Alamos National Laboratory, which could be documented in similar detail with sufficient time and effort.

Following the recent popularity of Large Language Models (LLMs), several attempts have been made to extend them to the visual domain. From having a visual assistant that could guide us through unfamiliar environments to generative models that produce images using only a high-level text description, the vision-language model (VLM) applications will significantly impact our relationship with technology. However, there are many challenges that need to be addressed to improve the reliability of those models. While language is discrete, vision evolves in a much higher dimensional space in which concepts cannot always be easily discretized. To better understand the mechanics behind mapping vision to language, we present this introduction to VLMs which we hope will help anyone who would like to enter the field. First, we introduce what VLMs are, how they work, and how to train them. Then, we present and discuss approaches to evaluate VLMs. Although this work primarily focuses on mapping images to language, we also discuss extending VLMs to videos.

Transversal gates are logical gate operations on encoded quantum information that are efficient in gate count and depth, and are designed to minimize error propagation. Efficient encoding circuits for quantum codes that admit transversal gates are thus crucial to reduce noise and realize useful quantum computers. The class of punctured Quantum Reed-Muller codes admit transversal gates. We construct resource efficient recursive encoders for the class of quantum codes constructed from Reed-Muller and punctured Reed-Muller codes. These encoders on $n$ qubits have circuit depth of $O(\log n)$ and lower gate counts compared to previous works. The number of CNOT gates in the encoder across bi-partitions of the qubits is found to be equal to the entanglement entropy across these partitions, demonstrating that the encoder is optimal in terms of CNOT gates across these partitions. Finally, connecting these ideas, we explicitly show that entanglement can be extracted from QRM codewords., Comment: 33 pages

Machine Learning (ML) based prognostics and health monitoring (PHM) tools provide new opportunities for manufacturers to operate and maintain their equipment in a risk-optimized manner and utilize it more sustainably along its lifecycle. Yet, in most industrial settings, data is often limited in quantity, and its quality can be inconsistent - both critical for developing and operating reliable ML models. To bridge this gap in practice, successfully industrialized PHM tools rely on the introduction of domain expertise as a prior, to enable sufficiently accurate predictions, while enhancing their interpretability. Thus, a key challenge while developing data-driven PHM tools involves translating the experience and process knowledge of maintenance personnel, development, and service engineers into a data structure. This structure must not only capture the diversity and variability of the expertise but also render this knowledge accessible for various data-driven algorithms. This results in data models that are heavily tailored towards a specific application and the failure modes the development team aims to detect or predict. The lack of a standardized approach limits developments' extensibility to new failure modes, their transferability to new applications, and it inhibits the utilization of standard data management and MLOps tools, increasing the burden on the development team. DeepFMEA draws inspiration from the Failure Mode and Effects Analysis (FMEA) in its structured approach to the analysis of any technical system and the resulting standardized data model, while considering aspects that are crucial to capturing process and maintenance expertise in a way that is both intuitive to domain experts and the resulting information can be introduced as priors to ML algorithms., Comment: 11 pages, 6 figures

We consider the question of how to best achieve the perception of eye contact when a person is captured by camera and then rendered on a 2D display. For single subjects photographed by a camera, conventional wisdom tells us that looking directly into the camera achieves eye contact. Through empirical user studies, we show that it is instead preferable to {\em look just below the camera lens}. We quantitatively assess where subjects should direct their gaze relative to a camera lens to optimize the perception that they are making eye contact.

Open Information Extraction (OIE) is a structured prediction (SP) task in Natural Language Processing (NLP) that aims to extract structured $n$-ary tuples - usually subject-relation-object triples - from free text. The word embeddings in the input text can be enhanced with linguistic features, usually Part-of-Speech (PoS) and Syntactic Dependency Parse (SynDP) labels. However, past enhancement techniques cannot leverage the power of pretrained language models (PLMs), which themselves have been hardly used for OIE. To bridge this gap, we are the first to leverage linguistic features with a Seq2Seq PLM for OIE. We do so by introducing two methods - Weighted Addition and Linearized Concatenation. Our work can give any neural OIE architecture the key performance boost from both PLMs and linguistic features in one go. In our settings, this shows wide improvements of up to 24.9%, 27.3% and 14.9% on Precision, Recall and F1 scores respectively over the baseline. Beyond this, we address other important challenges in the field: to reduce compute overheads with the features, we are the first ones to exploit Semantic Dependency Parse (SemDP) tags; to address flaws in current datasets, we create a clean synthetic dataset; finally, we contribute the first known study of OIE behaviour in SP models., Comment: Accepted at LREC-COLING 2024 Main Conference, Long Paper

Accurate phase estimation in the presence of unknown phase diffusive noise is a crucial yet challenging task in noisy quantum metrology. This problem is particularly interesting due to the detrimental impact of the associated noise. Here, we investigate the joint estimation of phase and phase diffusion using generalized Holland-Burnett states, known for their experimental accessibility. These states provide performance close to the optimal state in single-parameter phase estimation, even in the presence of photon losses. We adopt a twofold approach by analyzing the joint information extraction through the double homodyne measurement and the joint information availability across all probe states. Through our analysis, we find that the highest sensitivities are obtained by using states created by directing all input photons into one port of a balanced beam splitter. Furthermore, we infer that good levels of sensitivity persist even in the presence of moderate photon losses, illustrating the remarkable resilience of our probe states under lossy conditions., Comment: 11 pages, 5 figures

Quantum computers hold promise for solving problems intractable for classical computers, especially those with high time and/or space complexity. The reduction of the power flow (PF) problem into a linear system of equations, allows formulation of quantum power flow (QPF) algorithms, based on quantum linear system solving methods such as the Harrow-Hassidim-Lloyd (HHL) algorithm. The speedup due to QPF algorithms is claimed to be exponential when compared to classical PF solved by state-of-the-art algorithms. We investigate the potential for practical quantum advantage (PQA) in solving QPF compared to classical methods on gate-based quantum computers. We meticulously scrutinize the end-to-end complexity of QPF, providing a nuanced evaluation of the purported quantum speedup in this problem. Our analysis establishes a best-case bound for the HHL-QPF complexity, conclusively demonstrating the absence of any PQA in the direct current power flow (DCPF) and fast decoupled load flow (FDLF) problem. Additionally, we establish that for potential PQA to exist it is necessary to consider DCPF-type problems with a very narrow range of condition number values and readout requirements.

We introduce Lumos, the first end-to-end multimodal question-answering system with text understanding capabilities. At the core of Lumos is a Scene Text Recognition (STR) component that extracts text from first person point-of-view images, the output of which is used to augment input to a Multimodal Large Language Model (MM-LLM). While building Lumos, we encountered numerous challenges related to STR quality, overall latency, and model inference. In this paper, we delve into those challenges, and discuss the system architecture, design choices, and modeling techniques employed to overcome these obstacles. We also provide a comprehensive evaluation for each component, showcasing high quality and efficiency., Comment: Accepted to KDD 2024 (ADS Track)

Dynamic 2-[18F] fluoro-2-deoxy-D-glucose positron emission tomography (dFDG-PET) for human brain imaging has considerable clinical potential, yet its utilization remains limited. A key challenge in the quantitative analysis of dFDG-PET is characterizing a patient-specific blood input function, traditionally reliant on invasive arterial blood sampling. This research introduces a novel approach employing non-invasive deep learning model-based computations from the internal carotid arteries (ICA) with partial volume (PV) corrections, thereby eliminating the need for invasive arterial sampling. We present an end-to-end pipeline incorporating a 3D U-Net based ICA-net for ICA segmentation, alongside a Recurrent Neural Network (RNN) based MCIF-net for the derivation of a model-corrected blood input function (MCIF) with PV corrections. The developed 3D U-Net and RNN was trained and validated using a 5-fold cross-validation approach on 50 human brain FDG PET datasets. The ICA-net achieved an average Dice score of 82.18% and an Intersection over Union of 68.54% across all tested scans. Furthermore, the MCIF-net exhibited a minimal root mean squared error of 0.0052. The application of this pipeline to ground truth data for dFDG-PET brain scans resulted in the precise localization of seizure onset regions, which contributed to a successful clinical outcome, with the patient achieving a seizure-free state after treatment. These results underscore the efficacy of the ICA-net and MCIF-net deep learning pipeline in learning the ICA structure's distribution and automating MCIF computation with PV corrections. This advancement marks a significant leap in non-invasive neuroimaging.

We announce the initial release of "Airavata," an instruction-tuned LLM for Hindi. Airavata was created by fine-tuning OpenHathi with diverse, instruction-tuning Hindi datasets to make it better suited for assistive tasks. Along with the model, we also share the IndicInstruct dataset, which is a collection of diverse instruction-tuning datasets to enable further research for Indic LLMs. Additionally, we present evaluation benchmarks and a framework for assessing LLM performance across tasks in Hindi. Currently, Airavata supports Hindi, but we plan to expand this to all 22 scheduled Indic languages. You can access all artifacts at https://ai4bharat.github.io/airavata., Comment: Work in progress

This study addresses the challenge of extending Large Language Models (LLMs) to non-English languages that use non-Roman scripts. We propose an approach that utilizes the romanized form of text as an interface for LLMs, hypothesizing that its frequent informal use and shared tokens with English enhance cross-lingual alignment. Our approach involves the continual pretraining of an English LLM like Llama 2 on romanized text of non-English, non-Roman script languages, followed by instruction tuning on romanized data. The results indicate that romanized text not only reduces token fertility by 2x-4x but also matches or outperforms native script representation across various NLU, NLG, and MT tasks. Moreover, the embeddings computed on romanized text exhibit closer alignment with their English translations than those from the native script. Our approach presents a promising direction for leveraging the power of English LLMs in languages traditionally underrepresented in NLP. Our code is available on https://github.com/AI4Bharat/romansetu., Comment: Accepted to ACL 2024

The last few years have witnessed great success on image generation, which has crossed the acceptance thresholds of aesthetics, making it directly applicable to personal and commercial applications. However, images, especially in marketing and advertising applications, are often created as a means to an end as opposed to just aesthetic concerns. The goal can be increasing sales, getting more clicks, likes, or image sales (in the case of stock businesses). Therefore, the generated images need to perform well on these key performance indicators (KPIs), in addition to being aesthetically good. In this paper, we make the first endeavor to answer the question of "How can one infuse the knowledge of the end-goal within the image generation process itself to create not just better-looking images but also "better-performing'' images?''. We propose BoigLLM, an LLM that understands both image content and user behavior. BoigLLM knows how an image should look to get a certain required KPI. We show that BoigLLM outperforms 13x larger models such as GPT-3.5 and GPT-4 in this task, demonstrating that while these state-of-the-art models can understand images, they lack information on how these images perform in the real world. To generate actual pixels of behavior-conditioned images, we train a diffusion-based model (BoigSD) to align with a proposed BoigLLM-defined reward. We show the performance of the overall pipeline on two datasets covering two different behaviors: a stock dataset with the number of forward actions as the KPI and a dataset containing tweets with the total likes as the KPI, denoted as BoigBench. To advance research in the direction of utility-driven image generation and understanding, we release BoigBench, a benchmark dataset containing 168 million enterprise tweets with their media, brand account names, time of post, and total likes.

Realizing the recent advances in Natural Language Processing (NLP) to the legal sector poses challenging problems such as extremely long sequence lengths, specialized vocabulary that is usually only understood by legal professionals, and high amounts of data imbalance. The recent surge of Large Language Models (LLMs) has begun to provide new opportunities to apply NLP in the legal domain due to their ability to handle lengthy, complex sequences. Moreover, the emergence of domain-specific LLMs has displayed extremely promising results on various tasks. In this study, we aim to quantify how general LLMs perform in comparison to legal-domain models (be it an LLM or otherwise). Specifically, we compare the zero-shot performance of three general-purpose LLMs (ChatGPT-20b, LLaMA-2-70b, and Falcon-180b) on the LEDGAR subset of the LexGLUE benchmark for contract provision classification. Although the LLMs were not explicitly trained on legal data, we observe that they are still able to classify the theme correctly in most cases. However, we find that their mic-F1/mac-F1 performance is up to 19.2/26.8\% lesser than smaller models fine-tuned on the legal domain, thus underscoring the need for more powerful legal-domain LLMs., Comment: 7 pages, Accepted at Natural Legal Language Processing Workshop, EMNLP 2023

In this paper, we develop a modular neural network for real-time {\color{black}(> 10 Hz)} semantic mapping in uncertain environments, which explicitly updates per-voxel probabilistic distributions within a neural network layer. Our approach combines the reliability of classical probabilistic algorithms with the performance and efficiency of modern neural networks. Although robotic perception is often divided between modern differentiable methods and classical explicit methods, a union of both is necessary for real-time and trustworthy performance. We introduce a novel Convolutional Bayesian Kernel Inference (ConvBKI) layer which incorporates semantic segmentation predictions online into a 3D map through a depthwise convolution layer by leveraging conjugate priors. We compare ConvBKI against state-of-the-art deep learning approaches and probabilistic algorithms for mapping to evaluate reliability and performance. We also create a Robot Operating System (ROS) package of ConvBKI and test it on real-world perceptually challenging off-road driving data., Comment: arXiv admin note: text overlap with arXiv:2209.10663

In this paper, we present a model of a game among teams. Each team consists of a homogeneous population of agents. Agents within a team are cooperative while the teams compete with other teams. The dynamics and the costs are coupled through the empirical distribution (or the mean field) of the state of agents in each team. This mean-field is assumed to be observed by all agents. Agents have asymmetric information (also called a non-classical information structure). We propose a mean-field based refinement of the Team-Nash equilibrium of the game, which we call mean-field Markov perfect equilibrium (MF-MPE). We identify a dynamic programming decomposition to characterize MF-MPE. We then consider the case where each team has a large number of players and present a mean-field approximation which approximates the game among large-population teams as a game among infinite-population teams. We show that MF-MPE of the game among teams of infinite population is easier to compute and is an $\varepsilon$-approximate MF-MPE of the game among teams of finite population., Comment: 20 pages

With the advent of high-quality speech synthesis, there is a lot of interest in controlling various prosodic attributes of speech. Speaking rate is an essential attribute towards modelling the expressivity of speech. In this work, we propose a novel approach to control the speaking rate for non-autoregressive TTS. We achieve this by conditioning the speaking rate inside the duration predictor, allowing implicit speaking rate control. We show the benefits of this approach by synthesising audio at various speaking rate factors and measuring the quality of speaking rate-controlled synthesised speech. Further, we study the effect of the speaking rate distribution of the training data towards effective rate control. Finally, we fine-tune a baseline pretrained TTS model to obtain speaking rate control TTS. We provide various analyses to showcase the benefits of using this proposed approach, along with objective as well as subjective metrics. We find that the proposed methods have higher subjective scores and lower speaker rate errors across many speaking rate factors over the baseline., Comment: \c{opyright} 20XX IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works

Mott insulators, in which strong Coulomb interactions fully localize electrons on single atomic sites, play host to an incredibly rich and exciting array of strongly correlated physics. One can naturally extend this concept to cluster Mott insulators, wherein electrons localize not on single atoms but across clusters of atoms, forming ``molecules in solids''. The resulting localized degrees of freedom incorporate the full spectrum of electronic degrees of freedom, spin, orbital, and charge. These serve as the building blocks for cluster Mott insulators, and understanding them is an important first step toward understanding the many-body physics that emerges in candidate cluster Mott insulators. Here, we focus on elementary building blocks, neglecting some of the complexity present in real materials which can often obfuscate the underlying principles at play. Through an extensive set of exact theoretical calculations on clusters of varying geometry, number of orbitals, and number of electrons, we uncover some of the basic organizing principles of cluster Mott phases, particularly when interactions dominate and negate a simple single-particle picture. These include, for example, the identification of an additional ``cluster Hund's rule'', of cluster ground states that are best understood from a purely interacting perspective, and of several localized degrees of freedom which are protected by an unusual combination of discrete spatial or orbital symmetries. Finally, we discuss the impact of adding additional terms, relevant to material candidates, on the phase diagrams presented throughout, as well as the potential next steps in the journey to building a more complete picture of cluster Mott insulators., Comment: 18 pages, 23 figures, Appendix (1 page)

3D image reconstruction from a limited number of 2D images has been a long-standing challenge in computer vision and image analysis. While deep learning-based approaches have achieved impressive performance in this area, existing deep networks often fail to effectively utilize the shape structures of objects presented in images. As a result, the topology of reconstructed objects may not be well preserved, leading to the presence of artifacts such as discontinuities, holes, or mismatched connections between different parts. In this paper, we propose a shape-aware network based on diffusion models for 3D image reconstruction, named SADIR, to address these issues. In contrast to previous methods that primarily rely on spatial correlations of image intensities for 3D reconstruction, our model leverages shape priors learned from the training data to guide the reconstruction process. To achieve this, we develop a joint learning network that simultaneously learns a mean shape under deformation models. Each reconstructed image is then considered as a deformed variant of the mean shape. We validate our model, SADIR, on both brain and cardiac magnetic resonance images (MRIs). Experimental results show that our method outperforms the baselines with lower reconstruction error and better preservation of the shape structure of objects within the images., Comment: ShapeMI MICCAI 2023: Workshop on Shape in Medical Imaging

We present a general denoising algorithm for performing simultaneous tomography of quantum states and measurement noise. This algorithm allows us to fully characterize state preparation and measurement (SPAM) errors present in any quantum system. Our method is based on the analysis of the properties of the linear operator space induced by unitary operations. Given any quantum system with a noisy measurement apparatus, our method can output the quantum state and the noise matrix of the detector up to a single gauge degree of freedom. We show that this gauge freedom is unavoidable in the general case, but this degeneracy can be generally broken using prior knowledge on the state or noise properties, thus fixing the gauge for several types of state-noise combinations with no assumptions about noise strength. Such combinations include pure quantum states with arbitrarily correlated errors, and arbitrary states with block independent errors. This framework can further use available prior information about the setting to systematically reduce the number of observations and measurements required for state and noise detection. Our method effectively generalizes existing approaches to the problem, and includes as special cases common settings considered in the literature requiring an uncorrelated or invertible noise matrix, or specific probe states., Comment: 35 pages, 2 figures, Gauge transform definition fixed. DOI links fixed

The goal of our work is to investigate the oscillation and asymptotic properties of a class of difference equations with a condition. In contrast to most previous studies, the oscillation of the investigated equation is obtained with only one condition., Comment: 11 pages

Recent theoretical works on two-dimensional molybdenum disulfide, MoS$_2$, with sulfur vacancies predict that the suppression of thermal transport in MoS$_2$ by point defects is more prominent in monolayers and becomes negligible as layer number increases. Here, we investigate experimentally the thermal transport properties of two-dimensional molybdenum disulfide crystals with inherent sulfur vacancies. We study the first-order temperature coefficients of interlayer and intralayer Raman modes of MoS$_2$ crystals with different layer numbers and stacking orders. The in-plane thermal conductivity ($\kappa$) and total interface conductance per unit area ($ g $) across the 2D material-substrate interface of mono-, bi- and tri-layer MoS$_2$ samples are measured using the micro-Raman thermometry. Our results clearly demonstrate that the thermal conductivity is significantly suppressed by sulfur vacancies in monolayer MoS$_2$. However, this reduction in $\kappa$ becomes less evident as the layer number increases, confirming the theoretical predictions. No significant variation is observed in the $\kappa$ and $ g $ values of 2H and 3R stacked bilayer MoS$_2$ samples., Comment: Accepted for publication in Physical Review B; https://journals.aps.org/prb/accepted/7d078Of1Sb01894687c6532242ec71705e011e968

As Reinforcement Learning (RL) agents are increasingly employed in diverse decision-making problems using reward preferences, it becomes important to ensure that policies learned by these frameworks in mapping observations to a probability distribution of the possible actions are explainable. However, there is little to no work in the systematic understanding of these complex policies in a contrastive manner, i.e., what minimal changes to the policy would improve/worsen its performance to a desired level. In this work, we present COUNTERPOL, the first framework to analyze RL policies using counterfactual explanations in the form of minimal changes to the policy that lead to the desired outcome. We do so by incorporating counterfactuals in supervised learning in RL with the target outcome regulated using desired return. We establish a theoretical connection between Counterpol and widely used trust region-based policy optimization methods in RL. Extensive empirical analysis shows the efficacy of COUNTERPOL in generating explanations for (un)learning skills while keeping close to the original policy. Our results on five different RL environments with diverse state and action spaces demonstrate the utility of counterfactual explanations, paving the way for new frontiers in designing and developing counterfactual policies., Comment: ICML Workshop on Counterfactuals in Minds and Machines, 2023

The two-time scale nature of SAC, which is an actor-critic algorithm, is characterised by the fact that the critic estimate has not converged for the actor at any given time, but since the critic learns faster than the actor, it ensures eventual consistency between the two. Various strategies have been introduced in literature to learn better gradient estimates to help achieve better convergence. Since gradient estimates depend upon the critic, we posit that improving the critic can provide a better gradient estimate for the actor at each time. Utilizing this, we propose Soft Actor Retrospective Critic (SARC), where we augment the SAC critic loss with another loss term - retrospective loss - leading to faster critic convergence and consequently, better policy gradient estimates for the actor. An existing implementation of SAC can be easily adapted to SARC with minimal modifications. Through extensive experimentation and analysis, we show that SARC provides consistent improvement over SAC on benchmark environments. We plan to open-source the code and all experiment data at: https://github.com/sukritiverma1996/SARC., Comment: Accepted at RLDM 2022

Building a multilingual Automated Speech Recognition (ASR) system in a linguistically diverse country like India can be a challenging task due to the differences in scripts and the limited availability of speech data. This problem can be solved by exploiting the fact that many of these languages are phonetically similar. These languages can be converted into a Common Label Set (CLS) by mapping similar sounds to common labels. In this paper, new approaches are explored and compared to improve the performance of CLS based multilingual ASR model. Specific language information is infused in the ASR model by giving Language ID or using CLS to Native script converter on top of the CLS Multilingual model. These methods give a significant improvement in Word Error Rate (WER) compared to the CLS baseline. These methods are further tried on out-of-distribution data to check their robustness., Comment: 5 pages,5 figures, submitted to INTERSPEECH2023

Large Language Models (LLMs) have demonstrated impressive capabilities in creative tasks such as storytelling and E-mail generation. However, as LLMs are primarily trained on final text results rather than intermediate revisions, it might be challenging for them to perform text rewriting tasks. Most studies in the rewriting tasks focus on a particular transformation type within the boundaries of single sentences. In this work, we develop new strategies for instruction tuning and reinforcement learning to better align LLMs for cross-sentence rewriting tasks using diverse wording and structures expressed through natural languages including 1) generating rewriting instruction data from Wiki edits and public corpus through instruction generation and chain-of-thought prompting; 2) collecting comparison data for reward model training through a new ranking function. To facilitate this research, we introduce OpenRewriteEval, a novel benchmark covers a wide variety of rewriting types expressed through natural language instructions. Our results show significant improvements over a variety of baselines. The public repository is available on GitHub under Google Research (https://github.com/google-research/google-research/tree/master/rewritelm).

Explanation is a key component for the adoption of reinforcement learning (RL) in many real-world decision-making problems. In the literature, the explanation is often provided by saliency attribution to the features of the RL agent's state. In this work, we propose a complementary approach to these explanations, particularly for offline RL, where we attribute the policy decisions of a trained RL agent to the trajectories encountered by it during training. To do so, we encode trajectories in offline training data individually as well as collectively (encoding a set of trajectories). We then attribute policy decisions to a set of trajectories in this encoded space by estimating the sensitivity of the decision with respect to that set. Further, we demonstrate the effectiveness of the proposed approach in terms of quality of attributions as well as practical scalability in diverse environments that involve both discrete and continuous state and action spaces such as grid-worlds, video games (Atari) and continuous control (MuJoCo). We also conduct a human study on a simple navigation task to observe how their understanding of the task compares with data attributed for a trained RL policy. Keywords -- Explainable AI, Verifiability of AI Decisions, Explainable RL., Comment: Published at International Conference on Learning Representations (ICLR), 2023

Efficient representation of quantum many-body states on classical computers is a problem of enormous practical interest. An ideal representation of a quantum state combines a succinct characterization informed by the system's structure and symmetries, along with the ability to predict the physical observables of interest. A number of machine learning approaches have been recently used to construct such classical representations [1-6] which enable predictions of observables [7] and account for physical symmetries [8]. However, the structure of a quantum state gets typically lost unless a specialized ansatz is employed based on prior knowledge of the system [9-12]. Moreover, most such approaches give no information about what states are easier to learn in comparison to others. Here, we propose a new generative energy-based representation of quantum many-body states derived from Gibbs distributions used for modeling the thermal states of classical spin systems. Based on the prior information on a family of quantum states, the energy function can be specified by a small number of parameters using an explicit low-degree polynomial or a generic parametric family such as neural nets, and can naturally include the known symmetries of the system. Our results show that such a representation can be efficiently learned from data using exact algorithms in a form that enables the prediction of expectation values of physical observables. Importantly, the structure of the learned energy function provides a natural explanation for the hardness of learning for a given class of quantum states.

We present a catalog of likely astrophysical neutrino track-like events from the IceCube Neutrino Observatory. IceCube began reporting likely astrophysical neutrinos in 2016 and this system was updated in 2019. The catalog presented here includes events that were reported in real-time since 2019, as well as events identified in archival data samples starting from 2011. We report 275 neutrino events from two selection channels as the first entries in the catalog, the IceCube Event Catalog of Alert Tracks, which will see ongoing extensions with additional alerts. The gold and bronze alert channels respectively provide neutrino candidates with 50\% and 30\% probability of being astrophysical, on average assuming an astrophysical neutrino power law energy spectral index of 2.19. For each neutrino alert, we provide the reconstructed energy, direction, false alarm rate, probability of being astrophysical in origin, and likelihood contours describing the spatial uncertainty in the alert's reconstructed location. We also investigate a directional correlation of these neutrino events with gamma-ray and X-ray catalogs including 4FGL, 3HWC, TeVCat and Swift-BAT., Comment: Published in ApJS. Figure 4 and 6 corrected. Online version of the catalog is available on dataverse at https://doi.org/10.7910/DVN/SCRUCD

The LIGO/Virgo collaboration published the catalogs GWTC-1, GWTC-2.1 and GWTC-3 containing candidate gravitational-wave (GW) events detected during its runs O1, O2 and O3. These GW events can be possible sites of neutrino emission. In this paper, we present a search for neutrino counterparts of 90 GW candidates using IceCube DeepCore, the low-energy infill array of the IceCube Neutrino Observatory. The search is conducted using an unbinned maximum likelihood method, within a time window of 1000 s and uses the spatial and timing information from the GW events. The neutrinos used for the search have energies ranging from a few GeV to several tens of TeV. We do not find any significant emission of neutrinos, and place upper limits on the flux and the isotropic-equivalent energy emitted in low-energy neutrinos. We also conduct a binomial test to search for source populations potentially contributing to neutrino emission. We report a non-detection of a significant neutrino-source population with this test.

Dark Matter particles in the Galactic Center and halo can annihilate or decay into a pair of neutrinos producing a monochromatic flux of neutrinos. The spectral feature of this signal is unique and it is not expected from any astrophysical production mechanism. Its observation would constitute a dark matter smoking gun signal. We performed the first dedicated search with a neutrino telescope for such signal, by looking at both the angular and energy information of the neutrino events. To this end, a total of five years of IceCube's DeepCore data has been used to test dark matter masses ranging from 10~GeV to 40~TeV. No significant neutrino excess was found and upper limits on the annihilation cross section, as well as lower limits on the dark matter lifetime, were set. The limits reached are of the order of $10^{-24}$~cm$^3/s$ for an annihilation and up to $10^{27}$ seconds for decaying Dark Matter. Using the same data sample we also derive limits for dark matter annihilation or decay into a pair of Standard Model charged particles.

Trustera, the first functional system that redacts personally identifiable information (PII) in real-time spoken conversations to remove agents' need to hear sensitive information while preserving the naturalness of live customer-agent conversations. As opposed to post-call redaction, audio masking starts as soon as the customer begins speaking to a PII entity. This significantly reduces the risk of PII being intercepted or stored in insecure data storage. Trustera's architecture consists of a pipeline of automatic speech recognition, natural language understanding, and a live audio redactor module. The system's goal is three-fold: redact entities that are PII, mask the audio that goes to the agent, and at the same time capture the entity, so that the captured PII can be used for a payment transaction or caller identification. Trustera is currently being used by thousands of agents to secure customers' sensitive information., Comment: 5

We present the construction of standard entanglement-assisted (EA) qubit Reed-Muller (RM) codes and their tensor product variants from classical RM codes. We show that the EA RM codes obtained using the CSS construction have zero coding rate and negative catalytic rate. We further show that EA codes constructed from these same classical RM codes using the tensor product code (TPC) construction have positive coding rate and provide a subclass of EA RM TPCs that have positive catalytic rate, thus establishing the coding analog of superadditivity for this family of codes, useful towards quantum communications. We also generalize this analysis to obtain conditions for EA TPCs from classical codes to have positive catalytic rate when their corresponding EA CSS codes have zero rate., Comment: Minor changes: 21 pages, 3 figures. Accepted in Quantum

Atmospheric muon neutrinos are produced by meson decays in cosmic-ray-induced air showers. The flux depends on meteorological quantities such as the air temperature, which affects the density of air. Competition between decay and re-interaction of those mesons in the first particle production generations gives rise to a higher neutrino flux when the air density in the stratosphere is lower, corresponding to a higher temperature. A measurement of a temperature dependence of the atmospheric $\nu_{\mu}$ flux provides a novel method for constraining hadro\-nic interaction models of air showers. It is particularly sensitive to the production of kaons. Studying this temperature dependence for the first time requires a large sample of high-energy neutrinos as well as a detailed understanding of atmospheric properties. We report the significant ($> 10 \sigma$) observation of a correlation between the rate of more than 260,000 neutrinos, detected by IceCube between 2012 and 2018, and atmospheric temperatures of the stratosphere, measured by the Atmospheric Infrared Sounder (AIRS) instrument aboard NASA's AQUA satellite. For the observed 10$\%$ seasonal change of effective atmospheric temperature we measure a 3.5(3)$\%$ change in the muon neutrino flux. This observed correlation deviates by about 2-3 standard deviations from the expected correlation of 4.3$\%$ as obtained from theoretical predictions under the assumption of various hadronic interaction models, Comment: Preprint submitted to EPJC

Core-collapse supernovae are a promising potential high-energy neutrino source class. We test for correlation between seven years of IceCube neutrino data and a catalog containing more than 1000 core-collapse supernovae of types IIn and IIP and a sample of stripped-envelope supernovae. We search both for neutrino emission from individual supernovae, and for combined emission from the whole supernova sample through a stacking analysis. No significant spatial or temporal correlation of neutrinos with the cataloged supernovae was found. The overall deviation of all tested scenarios from the background expectation yields a p-value of 93% which is fully compatible with background. The derived upper limits on the total energy emitted in neutrinos are $1.7\times 10^{48}$ erg for stripped-envelope supernovae, $2.8\times 10^{48}$ erg for type IIP, and $1.3\times 10^{49}$ erg for type IIn SNe, the latter disfavouring models with optimistic assumptions for neutrino production in interacting supernovae. We conclude that strippe-envelope supernovae and supernovae of type IIn do not contribute more than $14.6\%$ and $33.9\%$ respectively to the diffuse neutrino flux in the energy range of about $10^3-10^5$ GeV, assuming that the neutrino energy spectrum follows a power-law with an index of $-2.5$. Under the same assumption, we can only constrain the contribution of type IIP SNe to no more than $59.9\%$. Thus core-collapse supernovae of types IIn and stripped-envelope supernovae can both be ruled out as the dominant source of the diffuse neutrino flux under the given assumptions., Comment: Submitted to ApJ Letters

Differentiating tumor progression (TP) from treatment-related necrosis (TN) is critical for clinical management decisions in glioblastoma (GBM). Dynamic FDG PET (dPET), an advance from traditional static FDG PET, may prove advantageous in clinical staging. dPET includes novel methods of a model-corrected blood input function that accounts for partial volume averaging to compute parametric maps that reveal kinetic information. In a preliminary study, a convolution neural network (CNN) was trained to predict classification accuracy between TP and TN for $35$ brain tumors from $26$ subjects in the PET-MR image space. 3D parametric PET Ki (from dPET), traditional static PET standardized uptake values (SUV), and also the brain tumor MR voxels formed the input for the CNN. The average test accuracy across all leave-one-out cross-validation iterations adjusting for class weights was $0.56$ using only the MR, $0.65$ using only the SUV, and $0.71$ using only the Ki voxels. Combining SUV and MR voxels increased the test accuracy to $0.62$. On the other hand, MR and Ki voxels increased the test accuracy to $0.74$. Thus, dPET features alone or with MR features in deep learning models would enhance prediction accuracy in differentiating TP vs TN in GBM.

Deep learning provides an excellent avenue for optimizing diagnosis and patient monitoring for clinical-based applications, which can critically enhance the response time to the onset of various conditions. For cardiovascular disease, one such condition where the rising number of patients increasingly outweighs the availability of medical resources in different parts of the world, a core challenge is the automated classification of various cardiac abnormalities. Existing deep learning approaches have largely been limited to detecting the existence of an irregularity, as in binary classification, which has been achieved using networks such as CNNs and RNN/LSTMs. The next step is to accurately perform multi-class classification and determine the specific condition(s) from the inherently noisy multi-variate waveform, which is a difficult task that could benefit from (1) a more powerful sequential network, and (2) the integration of clinical notes, which provide valuable semantic and clinical context from human doctors. Recently, Transformers have emerged as the state-of-the-art architecture for forecasting and prediction using time-series data, with their multi-headed attention mechanism, and ability to process whole sequences and learn both long and short-range dependencies. The proposed novel multi-modal Transformer architecture would be able to accurately perform this task while demonstrating the cross-domain effectiveness of Transformers, establishing a method for incorporating multiple data modalities within a Transformer for classification tasks, and laying the groundwork for automating real-time patient condition monitoring in clinical and ER settings., Comment: 18 pages, 11 figures, submitted to Artificial Intelligence in Medicine journal