Search

As robots become increasingly capable, users will want to describe high-level missions and have robots fill in the gaps. In many realistic settings, pre-built maps are difficult to obtain, so execution requires exploration and mapping that are necessary and specific to the mission. Consider an emergency response scenario where a user commands a robot, "triage impacted regions." The robot must infer relevant semantics (victims, etc.) and exploration targets (damaged regions) based on priors or other context, then explore and refine its plan online. These missions are incompletely specified, meaning they imply subtasks and semantics. While many semantic planning methods operate online, they are typically designed for well specified tasks such as object search or exploration. Recently, Large Language Models (LLMs) have demonstrated powerful contextual reasoning over a range of robotic tasks described in natural language. However, existing LLM planners typically do not consider online planning or complex missions; rather, relevant subtasks are provided by a pre-built map or a user. We address these limitations via SPINE (online Semantic Planner for missions with Incomplete Natural language specifications in unstructured Environments). SPINE uses an LLM to reason about subtasks implied by the mission then realizes these subtasks in a receding horizon framework. Tasks are automatically validated for safety and refined online with new observations. We evaluate SPINE in simulation and real-world settings. Evaluation missions require multiple steps of semantic reasoning and exploration in cluttered outdoor environments of over 20,000m$^2$ area. We evaluate SPINE against competitive baselines in single-agent and air-ground teaming applications. Please find videos and software on our project page: https://zacravichandran.github.io/SPINE

As usage of generative AI tools skyrockets, the amount of sensitive information being exposed to these models and centralized model providers is alarming. For example, confidential source code from Samsung suffered a data leak as the text prompt to ChatGPT encountered data leakage. An increasing number of companies are restricting the use of LLMs (Apple, Verizon, JPMorgan Chase, etc.) due to data leakage or confidentiality issues. Also, an increasing number of centralized generative model providers are restricting, filtering, aligning, or censoring what can be used. Midjourney and RunwayML, two of the major image generation platforms, restrict the prompts to their system via prompt filtering. Certain political figures are restricted from image generation, as well as words associated with women's health care, rights, and abortion. In our research, we present a secure and private methodology for generative artificial intelligence that does not expose sensitive data or models to third-party AI providers. Our work modifies the key building block of modern generative AI algorithms, e.g. the transformer, and introduces confidential and verifiable multiparty computations in a decentralized network to maintain the 1) privacy of the user input and obfuscation to the output of the model, and 2) introduce privacy to the model itself. Additionally, the sharding process reduces the computational burden on any one node, enabling the distribution of resources of large generative AI processes across multiple, smaller nodes. We show that as long as there exists one honest node in the decentralized computation, security is maintained. We also show that the inference process will still succeed if only a majority of the nodes in the computation are successful. Thus, our method offers both secure and verifiable computation in a decentralized network.

In this work, we take a deeper look into the diverse factors that influence the efficacy of modern unsupervised domain adaptation (UDA) methods using a large-scale, controlled empirical study. To facilitate our analysis, we first develop UDA-Bench, a novel PyTorch framework that standardizes training and evaluation for domain adaptation enabling fair comparisons across several UDA methods. Using UDA-Bench, our comprehensive empirical study into the impact of backbone architectures, unlabeled data quantity, and pre-training datasets reveals that: (i) the benefits of adaptation methods diminish with advanced backbones, (ii) current methods underutilize unlabeled data, and (iii) pre-training data significantly affects downstream adaptation in both supervised and self-supervised settings. In the context of unsupervised adaptation, these observations uncover several novel and surprising properties, while scientifically validating several others that were often considered empirical heuristics or practitioner intuitions in the absence of a standardized training and evaluation framework. The UDA-Bench framework and trained models are publicly available at https://github.com/ViLab-UCSD/UDABench_ECCV2024., Comment: ECCV 2024 Camera-ready version

Chalcogenide perovskites such as BaZrS3 are promising candidates for next generation optoelectronics such as photodetectors and solar cells. Compared to widely studied polycrystalline thin films, single crystals of BaZrS3 with minimal extended and point defects, are ideal platform to study the material's intrinsic transport properties and to make first-generation optoelectronic devices. However, the surface dielectrics formed on BaZrS3 single crystals due to sulfating or oxidation have led to significant challenges to achieving high quality electrical contacts, and hence, realizing the high-performance optoelectronic devices. Here, we report the development of electrical contact fabrication processes on BaZrS3 single crystals, where various processes were employed to address the surface dielectric issue. Moreover, with optimized electrical contacts fabricated through dry etching, high-performance BaZrS3 photoconductive devices with a low dark current of 0.1 nA at 10 V bias and a fast transient photoresponse with rise and decay time of <0.2 s were demonstrated. Our study emphasizes the importance of damage-free fabrication processes in making high-quality optoelectronic devices of BaZrS3 single crystals and sheds light on exploring its intrinsic transport properties.

For visual recognition, knowledge distillation typically involves transferring knowledge from a large, well-trained teacher model to a smaller student model. In this paper, we introduce an effective method to distill knowledge from an off-the-shelf vision-language model (VLM), demonstrating that it provides novel supervision in addition to those from a conventional vision-only teacher model. Our key technical contribution is the development of a framework that generates novel text supervision and distills free-form text into a vision encoder. We showcase the effectiveness of our approach, termed VLM-KD, across various benchmark datasets, showing that it surpasses several state-of-the-art long-tail visual classifiers. To our knowledge, this work is the first to utilize knowledge distillation with text supervision generated by an off-the-shelf VLM and apply it to vanilla randomly initialized vision encoders.

The dynamic collapse of pores under shock loading is thought to be directly related to hot spot generation and material failure, which is critical to the performance of porous energetic and structural materials. However, the shock compression response of porous materials at the local, individual pore scale is not well understood. This study examines, quantitatively, the collapse phenomenon of a single spherical void in PMMA at shock stresses ranging from 0.4-1.0 GPa. Using a newly developed internal digital image correlation technique in conjunction with plate impact experiments, full-field quantitative deformation measurements are conducted in the material surrounding the collapsing pore for the first time. The experimental results reveal two failure mode transitions as shock stress is increased: (i) the first in-situ evidence of shear localization via adiabatic shear banding and (ii) dynamic fracture initiation at the pore surface. Numerical simulations using thermo-viscoplastic dynamic finite element analysis provide insights into the formation of adiabatic shear bands (ASBs) and stresses at which failure mode transitions occur. Further numerical and theoretical modeling indicates the dynamic fracture to occur along the weakened material inside an adiabatic shear band. Finally, analysis of the evolution of pore asymmetry and models for ASB spacing elucidate the mechanisms for the shear band initiation sites, and elastostatic theory explains the experimentally observed ASB and fracture paths based on the directions of maximum shear.

Emission spectra and diagnostic spectral features of a diverse range of ablated meteorite samples with a known composition are presented. We aim to provide a reference spectral dataset to improve our abilities to classify meteoroid composition types from meteor spectra observations. The data were obtained by ablating meteorite samples in high-enthalpy plasma wind tunnel facilities recreating conditions characteristic of low-speed meteors. Near-UV to visible-range (320 - 800 nm) emission spectra of 22 diverse meteorites captured by a high-resolution Echelle spectrometer were analyzed to identify the characteristic spectral features of individual meteorite groups. The same dataset captured by a lower-resolution meteor spectrograph was applied to compare the meteorite data with meteor spectra observations. Spectral modeling revealed that the emitting meteorite plasma was characterized by temperatures of 3700 - 4800 K, similar to the main temperature component of meteors. The studied line intensity variations were found to trace the differences in the original meteorite composition and thus can be used to constrain the individual meteorite classes. We demonstrate that meteorite composition types, including ordinary chondrites, carbonaceous chondrites, various achondrites, stony-iron and iron meteorites, can be spectrally distinguished by measuring relative line intensities of Mg I, Fe I, Na I, Cr I, Mn I, Si I, H I, CN, Ni I, and Li I. Additionally, we confirm the effect of the incomplete evaporation of refractory elements Al, Ti, and Ca, and the presence of minor species Co I, Cu I, and V I.

We present a general fabrication strategy for freestanding single-crystalline complex oxide device arrays via wet chemical etching-based microfabrication processes and epitaxial lift-off. Here, we used 0.5Ba(Zr$_{0.2}$Ti$_{0.8}$)O$_3$-0.5Ba(Zr$_{0.7}$Ti$_{0.3}$)O$_3$ (BCZT) as a model relaxor ferroelectric oxide system and La$_{0.7}$Sr$_{0.3}$MnO$_3$ as the sacrificial layer for demonstration. Arrays of SrRuO$_3$ (SRO) / BCZT / SRO ferroelectric capacitor mesas were first defined and isolated on the growth wafer, and then they were released using epitaxial lift-off with lithography-defined surrounding etching holes, after which the freestanding device arrays were integrated onto a glass substrate. Our proposed strategy sheds light on preparing various freestanding single-crystalline oxide devices and paves the way for their heterogeneous integration onto arbitrary substates.

Noncollinear ferroic materials are sought after as testbeds to explore the intimate connections between topology and symmetry, which result in electronic, optical and magnetic functionalities not observed in collinear ferroic materials. For example, ferroaxial materials have ordered rotational structural distortions that break mirror symmetry and induce chirality. When ferroaxial order is coupled with ferroelectricity arising from a broken inversion symmetry, it offers the prospect of electric-field-control of the ferroaxial distortions and opens up new tunable functionalities. However, chiral ferroelectrics, especially ones stable at room temperature, are rare. We report the discovery of a strain-stabilized, room-temperature chiral ferroelectric phase in single crystals of BaTiS$_3$, a quasi-one-dimensional (1D) hexagonal chalcogenide. Using first-principles calculations, we predict the stabilization of this multiferroic phase having $P6_3$ space group for biaxial tensile strains exceeding 1.5% applied on the basal ab-plane of the room temperature $P6_3cm$ phase of BaTiS$_3$. The chiral ferroelectric phase is characterized by rotational distortions of select TiS$_6$ octahedra around the long $c$-axis and polar displacement of Ti atoms along the $c$-axis. We used an innovative approach using focused ion beam milling to make appropriately strained samples of BaTiS$_3$. The ferroaxial and ferroelectric distortions, and their domains in $P6_3$-BaTiS$_3$ were directly resolved using atomic resolution scanning transmission electron microscopy. Landau-based phenomenological modeling predicts a strong coupling between the ferroelectric and the ferroaxial order making $P6_3$-BaTiS$_3$ an attractive test bed for achieving electric-field control of chirality-related phenomena such as circular photo-galvanic current and the Rashba effect., Comment: 19 pages, 4 figures

We are often interested in decomposing complex, structured data into simple components that explain the data. The linear version of this problem is well-studied as dictionary learning and factor analysis. In this work, we propose a combinatorial model in which to study this question, motivated by the way objects occlude each other in a scene to form an image. First, we identify a property we call "well-structuredness" of a set of low-dimensional components which ensures that no two components in the set are too similar. We show how well-structuredness is sufficient for learning the set of latent components comprising a set of sample instances. We then consider the problem: given a set of components and an instance generated from some unknown subset of them, identify which parts of the instance arise from which components. We consider two variants: (1) determine the minimal number of components required to explain the instance; (2) determine the correct explanation for as many locations as possible. For the latter goal, we also devise a version that is robust to adversarial corruptions, with just a slightly stronger assumption on the components. Finally, we show that the learning problem is computationally infeasible in the absence of any assumptions., Comment: 31 pages, 3 figures

Ferroelectricity is characterized by the presence of spontaneous and switchable macroscopic polarization. Scaling limits of ferroelectricity have been of both fundamental and technological importance, but the probes of ferroelectricity have often been indirect due to confounding factors such as leakage in the direct electrical measurements. Recent interest in low-voltage switching electronic devices squarely puts the focus on ultrathin limits of ferroelectricity in an electronic device form, specifically on the robustness of ferroelectric characteristics such as retention and endurance for practical applications. Here, we illustrate how manipulating the kinetic energy of the plasma plume during pulsed laser deposition can yield ultrathin ferroelectric capacitor heterostructures with high bulk and interface quality, significantly low leakage currents and a broad "growth window". These heterostructures venture into previously unexplored aspects of ferroelectric properties, showcasing ultralow switching voltages ($<$0.3 V), long retention times ($>$10$^{4}$s), and high endurance ($>$10$^{11}$cycles) in 20 nm films of the prototypical perovskite ferroelectric, BaTiO$_{3}$. Our work demonstrates that materials engineering can push the envelope of performance for ferroelectric materials and devices at the ultrathin limit and opens a direct, reliable and scalable pathway to practical applications of ferroelectrics in ultralow voltage switches for logic and memory technologies.

We introduce InVi, an approach for inserting or replacing objects within videos (referred to as inpainting) using off-the-shelf, text-to-image latent diffusion models. InVi targets controlled manipulation of objects and blending them seamlessly into a background video unlike existing video editing methods that focus on comprehensive re-styling or entire scene alterations. To achieve this goal, we tackle two key challenges. Firstly, for high quality control and blending, we employ a two-step process involving inpainting and matching. This process begins with inserting the object into a single frame using a ControlNet-based inpainting diffusion model, and then generating subsequent frames conditioned on features from an inpainted frame as an anchor to minimize the domain gap between the background and the object. Secondly, to ensure temporal coherence, we replace the diffusion model's self-attention layers with extended-attention layers. The anchor frame features serve as the keys and values for these layers, enhancing consistency across frames. Our approach removes the need for video-specific fine-tuning, presenting an efficient and adaptable solution. Experimental results demonstrate that InVi achieves realistic object insertion with consistent blending and coherence across frames, outperforming existing methods.

Multimodal synthetic data generation is crucial in domains such as autonomous driving, robotics, augmented/virtual reality, and retail. We propose a novel approach, GenMM, for jointly editing RGB videos and LiDAR scans by inserting temporally and geometrically consistent 3D objects. Our method uses a reference image and 3D bounding boxes to seamlessly insert and blend new objects into target videos. We inpaint the 2D Regions of Interest (consistent with 3D boxes) using a diffusion-based video inpainting model. We then compute semantic boundaries of the object and estimate it's surface depth using state-of-the-art semantic segmentation and monocular depth estimation techniques. Subsequently, we employ a geometry-based optimization algorithm to recover the 3D shape of the object's surface, ensuring it fits precisely within the 3D bounding box. Finally, LiDAR rays intersecting with the new object surface are updated to reflect consistent depths with its geometry. Our experiments demonstrate the effectiveness of GenMM in inserting various 3D objects across video and LiDAR modalities.

Topological defects, such as vortices and skyrmions in magnetic and dipolar systems, can give rise to properties that are not observed in typical magnets or dielectrics. Here, we report the discovery of an atomic-scale dipolar vortex lattice in the charge-density-wave (CDW) phase of BaTiS3, a quasi-one-dimensional (quasi-1D) hexagonal chalcogenide, using X-ray synchrotron single-crystal diffraction studies. The vortex lattice consists of a periodic array of vortex-vortex-antivortex patterns composed of electric dipoles from off-center displacements of octahedrally coordinated Ti atoms. Using first-principles calculations and phenomenological modeling, we show that the dipolar vortex lattice in BaTiS3 arises from the coupling between multiple lattice instabilities arising from flat, soft phonon bands. This mechanism contrasts with classical dipolar textures in ferroelectric heterostructures that emerge from the competition between electrostatic and strain energies, and necessitate a dimensional reduction in the form of thin films and heterostructures to stabilize the textures. The observation of dipolar vortices in BaTiS3 brings the ultimate scaling limit for dipolar topologies down to about a nanometer and unveils the intimate connection between crystal symmetry and real-space topology. Our work sets up zero-filling triangular lattice materials with instabilities as a playground for realizing and understanding quantum polarization topologies.

Low-weight, passive, thermal-adaptive radiation technologies are needed to maintain an operable temperature for spacecraft while they experience various energy fluxes. In this study, we used a thin-film coating with the Fabry-Perot (FP) effect to enhance emissivity contrast ({\Delta}{\epsilon}) between VO2 phase-change states. This coating utilizes a novel hybrid material architecture that combines VO2 with a mid- and long-wave infrared transparent chalcogenide, zinc sulfide (ZnS), as a cavity spacer layer. We simulated the design parameter space to obtain a theoretical maximum {\Delta}{\epsilon} of 0.63 and grew prototype devices. Using X-ray diffraction, Raman spectroscopy, and Fourier Transform Infrared (FTIR) Spectroscopy, we determined that an intermediate buffer layer of TiO2 is necessary to execute the crystalline growth of monoclinic VO2 on ZnS. Through temperature-dependent FTIR spectroscopy measurements, our fabricated devices demonstrated FP-cavity enhanced adaptive thermal emittance., Comment: 16 pages, 5 figures, full article

Neural networks that can capture key principles underlying brain computation offer exciting new opportunities for developing artificial intelligence and brain-like computing algorithms. Such networks remain biologically plausible while leveraging localized forms of synaptic learning rules and modular network architecture found in the neocortex. Compared to backprop-driven deep learning approches, they provide more suitable models for deploying on neuromorphic hardware and have greater potential for scalability on large-scale computing clusters. The development of such brain-like neural networks depends on having a learning procedure that can build effective internal representations from data. In this work, we introduce and evaluate a brain-like neural network model capable of unsupervised representation learning. It builds on the Bayesian Confidence Propagation Neural Network (BCPNN), which has earlier been implemented as abstract as well as biophyscially detailed recurrent attractor neural networks explaining various cortical associative memory phenomena. Here we developed a feedforward BCPNN model to perform representation learning by incorporating a range of brain-like attributes derived from neocortical circuits such as cortical columns, divisive normalization, Hebbian synaptic plasticity, structural plasticity, sparse activity, and sparse patchy connectivity. The model was tested on a diverse set of popular machine learning benchmarks: grayscale images (MNIST, Fashion-MNIST), RGB natural images (SVHN, CIFAR-10), QSAR (MUV, HIV), and malware detection (EMBER). The performance of the model when using a linear classifier to predict the class labels fared competitively with conventional multi-layer perceptrons and other state-of-the-art brain-like neural networks.

In this paper, we relate the philosophical literature on pessimism traps to information cascades, a formal model derived from the economics and mathematics literature. A pessimism trap is a social pattern in which individuals in a community, in situations of uncertainty, begin to copy the sub-optimal actions of others, despite their individual beliefs. This maps nicely onto the concept of an information cascade, which involves a sequence of agents making a decision between two alternatives, with a private signal of the superior alternative and a public history of others' actions. Key results from the economics literature show that information cascades occur with probability one in many contexts, and depending on the strength of the signal, populations can fall into the incorrect cascade very easily and quickly. Once formed, in the absence of external perturbation, a cascade cannot be broken -- therefore, we derive an intervention that can be used to nudge a population from an incorrect to a correct cascade and, importantly, maintain the cascade once the subsidy is discontinued. We study this both theoretically and empirically., Comment: 10 pages, 5 figures

Networks of interconnected neurons communicating through spiking signals offer the bedrock of neural computations. Our brains spiking neural networks have the computational capacity to achieve complex pattern recognition and cognitive functions effortlessly. However, solving real-world problems with artificial spiking neural networks (SNNs) has proved to be difficult for a variety of reasons. Crucially, scaling SNNs to large networks and processing large-scale real-world datasets have been challenging, especially when compared to their non-spiking deep learning counterparts. The critical operation that is needed of SNNs is the ability to learn distributed representations from data and use these representations for perceptual, cognitive and memory operations. In this work, we introduce a novel SNN that performs unsupervised representation learning and associative memory operations leveraging Hebbian synaptic and activity-dependent structural plasticity coupled with neuron-units modelled as Poisson spike generators with sparse firing (~1 Hz mean and ~100 Hz maximum firing rate). Crucially, the architecture of our model derives from the neocortical columnar organization and combines feedforward projections for learning hidden representations and recurrent projections for forming associative memories. We evaluated the model on properties relevant for attractor-based associative memories such as pattern completion, perceptual rivalry, distortion resistance, and prototype extraction.

The ethical integration of Artificial Intelligence (AI) in healthcare necessitates addressing fairness-a concept that is highly context-specific across medical fields. Extensive studies have been conducted to expand the technical components of AI fairness, while tremendous calls for AI fairness have been raised from healthcare. Despite this, a significant disconnect persists between technical advancements and their practical clinical applications, resulting in a lack of contextualized discussion of AI fairness in clinical settings. Through a detailed evidence gap analysis, our review systematically pinpoints several deficiencies concerning both healthcare data and the provided AI fairness solutions. We highlight the scarcity of research on AI fairness in many medical domains where AI technology is increasingly utilized. Additionally, our analysis highlights a substantial reliance on group fairness, aiming to ensure equality among demographic groups from a macro healthcare system perspective; in contrast, individual fairness, focusing on equity at a more granular level, is frequently overlooked. To bridge these gaps, our review advances actionable strategies for both the healthcare and AI research communities. Beyond applying existing AI fairness methods in healthcare, we further emphasize the importance of involving healthcare professionals to refine AI fairness concepts and methods to ensure contextually relevant and ethically sound AI applications in healthcare.

Hydrogen has been identified as a clean, zero carbon, sustainable, and promising energy source for the future, and electrochemical water splitting for hydrogen production is an emission-free, efficient energy conversion technology. A major limitation of this approach is the unavailability of efficient, abundant, inexpensive catalysts, which prompts the need for new catalytic materials. Here, we report the synthesis and electrocatalytic properties of a novel transition metal-based ternary chalcogenide family, LaMS$_3$ (M = Mn, Fe, Co, Ni). Powder X-ray diffraction confirms the phase purity of these materials, while composition analysis using energy dispersive spectroscopy (EDS) confirms the presence of the stoichiometric ratio of elements in these compounds. X-ray photoelectron spectroscopy (XPS) and X-ray absorption spectroscopy (XAS) were used to study the chemical states on the surface and in bulk, respectively. These materials exhibit bifunctional catalytic activity towards the two half-reactions of the water-splitting process, with LaNiS$_3$ being the most active material for both hydrogen evolution reaction (HER) and oxygen evolution reaction (OER). The LaMS$_3$ compounds show long-term stability with negligible change in the overpotential at a constant current density of 10 mA cm$^{-2}$ over 18 hours of measurements. As compared to the corresponding ternary oxides, the LaMS$_3$ materials exhibit higher activity and significantly lower Tafel slopes. The ability to catalyze both half-reactions of water electrolysis makes these materials promising candidates for bifunctional catalysts and presents a new avenue to search for high-efficiency electrocatalysts for water splitting.

One common and desirable application of robots is exploring potentially hazardous and unstructured environments. Air-ground collaboration offers a synergistic approach to addressing such exploration challenges. In this paper, we demonstrate a system for large-scale exploration using a team of aerial and ground robots. Our system uses semantics as lingua franca, and relies on fully opportunistic communications. We highlight the unique challenges from this approach, explain our system architecture and showcase lessons learned during our experiments. All our code is open-source, encouraging researchers to use it and build upon., Comment: 6 pages, 5 figres

A home-built compact system that functions as both a transfer stage for deterministic transfer processes and a mask aligner for contact photolithography, is constructed. The precision translation sample stage and optical microscope are shared between the two modes. In the transfer mode, assisted by either an adhesive or a heating element, the setup has been used to deterministically transfer freestanding semiconductors, 2D materials, and van der Waals electrodes. When configured in photolithography mode, the proposed instrument has been employed to fabricate various micro-scale patterns and devices, with minimum feature sizes of ~1-2 um achieved. Our prototype instrument provides a feasible solution for preforming high quality deterministic transfer and photolithography processes on one single tool in house.

The ISMR profoundly impacts over a billion people across the region. ISMR extremes have been linked to the ENSO and modulated by the Indian Ocean Dipole (IOD). ISMR of 2022 displayed intriguing spatial patterns: above-normal precipitation over the south peninsula and central India, normal over Northwest India, and below-normal over East and NE India. In 2022, La Nina conditions associated with a negative IOD (nIOD), were prevalent over the equatorial Pacific Ocean (PO) and IO. This study investigates the often-overlooked Oceanic subsurface contribution to large-scale atmospheric convection over the tropical IO. By undertaking a comprehensive analysis, we find that strong equatorial westerly wind anomalies prevailed over the equatorial IO during co-occurring years generating eastward propagating downwelling Kelvin waves which deepens the D20 in the eastern equatorial IO. The Kelvin waves then propagates into BoB and gets reflected from southern tip of India as westward propagating Rossby waves deepening the D20 and causing low level wind convergence in the Northern Arabian Sea(NAS). We identify two centres of low-level wind convergence, one over NAS and another over Maritime Continent during co-occurrence years. These wind convergence centres play a pivotal role in channeling moisture towards their core, also, significantly deepen D20, increase the Ocean heat content and thus maintaining warmer SSTs over area of conversions. Together, these interrelated factors create an environment that is conducive for enhanced convective activity over the zone of convergence, enhancing and sustaining the convective conditions. This study underscores the significant contribution of ocean dynamics in shaping large-scale atmospheric convection during the co-occurrence years, highlighting the significance of considering broader Oceanic variables in comprehending and forecasting monsoonal variability., Comment: 47 pages, 10 main figures, 08 supplementary figures; under review in Climate Dynamics

We give nearly-tight upper and lower bounds for the improving multi-armed bandits problem. An instance of this problem has $k$ arms, each of whose reward function is a concave and increasing function of the number of times that arm has been pulled so far. We show that for any randomized online algorithm, there exists an instance on which it must suffer at least an $\Omega(\sqrt{k})$ approximation factor relative to the optimal reward. We then provide a randomized online algorithm that guarantees an $O(\sqrt{k})$ approximation factor, if it is told the maximum reward achievable by the optimal arm in advance. We then show how to remove this assumption at the cost of an extra $O(\log k)$ approximation factor, achieving an overall $O(\sqrt{k} \log k)$ approximation relative to optimal., Comment: 12 pages, 0 figures

Recent advances in machine learning have demonstrated that multi-modal pre-training can improve automatic speech recognition (ASR) performance compared to randomly initialized models, even when models are fine-tuned on uni-modal tasks. Existing multi-modal pre-training methods for the ASR task have primarily focused on single-stage pre-training where a single unsupervised task is used for pre-training followed by fine-tuning on the downstream task. In this work, we introduce a novel method combining multi-modal and multi-task unsupervised pre-training with a translation-based supervised mid-training approach. We empirically demonstrate that such a multi-stage approach leads to relative word error rate (WER) improvements of up to 38.45% over baselines on both Librispeech and SUPERB. Additionally, we share several important findings for choosing pre-training methods and datasets., Comment: Accepted in LREC-COLING 2024 - The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation

BaTiS3, a quasi-1D complex chalcogenide, has gathered considerable scientific and technological interest due to its giant optical anisotropy and electronic phase transitions. However, the synthesis of high-quality BaTiS3 crystals, particularly those featuring crystal sizes of millimeters or larger, remains a challenge. Here, we investigate the growth of BaTiS3 crystals utilizing a molten salt flux of either potassium iodide, or a mixture of barium chloride and barium iodide. The crystals obtained through this method exhibit a substantial increase in volume compared to those synthesized via the chemical vapor transport method, while preserving their intrinsic optical and electronic properties. Our flux growth method provides a promising route towards the production of high-quality, large-scale single crystals of BaTiS3, which will greatly facilitate advanced characterizations of BaTiS3 and its practical applications that require large crystal dimensions. Additionally, our approach offers an alternative synthetic route for other emerging complex chalcogenides.

In the rapidly evolving landscape of artificial intelligence (AI), the collaboration between human intelligence and AI systems, known as Human-AI (HAI) Teaming, has emerged as a cornerstone for advancing problem-solving and decision-making processes. The advent of Large Pre-trained Models (LPtM) has significantly transformed this landscape, offering unprecedented capabilities by leveraging vast amounts of data to understand and predict complex patterns. This paper surveys the pivotal integration of LPtMs with HAI, emphasizing how these models enhance collaborative intelligence beyond traditional approaches. It examines the potential of LPtMs in augmenting human capabilities, discussing this collaboration for AI model improvements, effective teaming, ethical considerations, and their broad applied implications in various sectors. Through this exploration, the study sheds light on the transformative impact of LPtM-enhanced HAI Teaming, providing insights for future research, policy development, and strategic implementations aimed at harnessing the full potential of this collaboration for research and societal benefit.

A key open question in the study of layered superconducting nickelate films is the role that hydrogen incorporation into the lattice plays in the appearance of the superconducting state. Due to the challenges of stabilizing highly crystalline square planar nickelate films, films are prepared by the deposition of a more stable parent compound which is then transformed into the target phase via a topotactic reaction with a strongly reducing agent such as CaH$_2$. Recent studies, both experimental and theoretical, have introduced the possibility that the incorporation of hydrogen from the reducing agent into the nickelate lattice may be critical for the superconductivity. In this work, we use secondary ion mass spectrometry to examine superconducting La$_{1-x}$X$_x$NiO$_2$ / SrTiO$_3$ (X = Ca and Sr) and Nd$_6$Ni$_5$O$_{12}$ / NdGaO$_3$ films, along with non-superconducting NdNiO$_2$ / SrTiO$_3$ and (Nd,Sr)NiO$_2$ / SrTiO$_3$. We find no evidence for extensive hydrogen incorporation across a broad range of samples, including both superconducting and non-superconducting films. Theoretical calculations indicate that hydrogen incorporation is broadly energetically unfavorable in these systems, supporting our conclusion that hydrogen incorporation is not generally required to achieve a superconducting state in layered square-planar nickelates., Comment: 23 pages, 21 figures (including supplemental information)

We report a comparative study of temperature-dependent photoluminescence and structural dynamics of two perovskite semiconductors, the chalcogenide BaZrS$_3$ (BZS) and the halide CsPbBr$_3$ (CPB). These materials have similar crystal structures and direct band gaps, but we find that they have quite distinct optoelectronic and vibrational properties. Both materials exhibit thermally-activated non-radiative recombination, but the non-radiative recombination rate in BZS is between two and four orders of magnitude faster than in CPB. Raman spectroscopy reveals that the effects of phonon anharmonicity are far more pronounced in CPB than in BZS. Further, although both materials feature a large dielectric response due to low-energy polar optical phonons, the phonons in CPB are substantially lower in energy than in BZS. Our results suggest that electron-phonon coupling in BZS is more effective at non-radiative recombination than in CPB, and that BZS may also have a substantially higher concentration of non-radiative recombination centers than CPB. The low defect concentration in CPB may be related to the ease of lattice reconfiguration, typified by anharmonic bonding. It remains to be seen to what extent these differences are inherent to the chalcogenide and halide perovskites and to what extent they can be affected by materials processing; comparing BZS single-crystals and thin films provides reason for optimism., Comment: Main text - 12 pages with 5 figures and 1 table. Supplemental text - 16 pages with 6 figures and 5 tables

This paper studies the evolving domain of Continual Learning (CL) in large language models (LLMs), with a focus on developing strategies for efficient and sustainable training. Our primary emphasis is on continual domain-adaptive pretraining, a process designed to equip LLMs with the ability to integrate new information from various domains while retaining previously learned knowledge and enhancing cross-domain knowledge transfer without relying on domain-specific identification. Unlike previous studies, which mostly concentrate on a limited selection of tasks or domains and primarily aim to address the issue of forgetting, our research evaluates the adaptability and capabilities of LLMs to changing data landscapes in practical scenarios. To this end, we introduce a new benchmark designed to measure the adaptability of LLMs to these evolving data environments, offering a comprehensive framework for evaluation. We examine the impact of model size on learning efficacy and forgetting, as well as how the progression and similarity of emerging domains affect the knowledge transfer within these models. Our findings uncover several key insights: (i) when the sequence of domains shows semantic similarity, continual pretraining enables LLMs to better specialize in the current domain compared to stand-alone fine-tuning, (ii) training across a diverse range of domains enhances both backward and forward knowledge transfer, and (iii) smaller models are particularly sensitive to continual pretraining, showing the most significant rates of both forgetting and learning. We posit that our research marks a shift towards establishing a more realistic benchmark for investigating CL in LLMs, and has the potential to play a key role in guiding the direction of future research in the field.

This paper explores a novel technique for improving recall in cross-language information retrieval (CLIR) systems using iterative query refinement grounded in the user's lexical-semantic space. The proposed methodology combines multi-level translation, semantic embedding-based expansion, and user profile-centered augmentation to address the challenge of matching variance between user queries and relevant documents. Through an initial BM25 retrieval, translation into intermediate languages, embedding lookup of similar terms, and iterative re-ranking, the technique aims to expand the scope of potentially relevant results personalized to the individual user. Comparative experiments on news and Twitter datasets demonstrate superior performance over baseline BM25 ranking for the proposed approach across ROUGE metrics. The translation methodology also showed maintained semantic accuracy through the multi-step process. This personalized CLIR framework paves the path for improved context-aware retrieval attentive to the nuances of user language., Comment: This is just an initial idea and it's implementation. The results are computed for the first 100 data points. Detailed results will be published with the actual paper

Rare earth dopants are one of the most extensively studied optical emission centers for a broad range of applications such as laser optoelectronics, sensing, lighting, and quantum information technologies due to their narrow optical linewidth and exceptional coherence properties. Epitaxial doped oxide thin films can serve as a promising and controlled host to investigate rare-earth dopants suitable for scalable quantum memories, on-chip lasers and amplifiers. Here, we report high-quality epitaxial thin films of Tm-doped CaZrO$_3$ grown by pulsed laser deposition for infrared optoelectronic and quantum memory applications. We perform extensive structural and chemical characterization to probe the crystallinity of the films and the doping behavior. Low temperature photoluminescence measurements show sharp radiative transitions in the short-wave infrared range of 1.75 - 2 \mu m., Comment: 17 pages, 4 figures

Polarimetric infrared detection bolsters IR thermography by leveraging the polarization of light. Optical anisotropy, i.e., birefringence and dichroism, can be leveraged to achieve polarimetric detection. Recently, giant optical anisotropy was discovered in quasi-1D narrow-bandgap hexagonal perovskite sulfides, A1+xTiS3, specifically BaTiS3[1,2] and Sr9/8TiS3[3,4]. In these materials, the critical role of atomic-scale structure modulations[4,5] in the unconventional electrical[5,6], optical[7,8], and thermal[7,9] properties raises the broader question of other materials that belong to this family. To address this issue, for the first time, we synthesized high-quality single crystals of a largely unexplored member of the A1+xTiX3 (X = S, Se) family, BaTiSe3. Single-crystal X-ray diffraction determined the room-temperature structure with the P31c space group, which is a superstructure of the earlier reported[10] P63/mmc structure. The crystal structure of BaTiSe3 features antiparallel c-axis displacements similar to BaTiS3,[2] but is of lower symmetry. Polarization-resolved Raman and Fourier transform infrared (FTIR) spectroscopy were used to characterize the optical anisotropy of BaTiSe3, whose refractive index along the ordinary (perpendicular to c) and extraordinary (parallel to c) optical axes was quantitatively determined by combining ellipsometry studies with FTIR. With a giant birefringence {\Delta}n~0.9, BaTiSe3 emerges as a new candidate for miniaturized birefringent optics for mid-wave infrared to long-wave infrared imaging.