Search

Magnons carriers of spin information can be controlled by electric fields in the multiferroic BiFeO$_3$ (BFO), a milestone that brings magnons closer to application in future devices. The origin of magnon-spin currents in BFO, however, is not fully understood due to BFO's complicated magnetic texture. In this letter, we present a phenomenological model to elucidate the existence of magnon spin currents in generalized multiferroics by examining the symmetries inherent to their magnetic and polar structures. This model is grounded in experimental data obtained from BFO and its derivatives, which informs the symmetry operations and resultant magnon behavior. By doing so, we address the issue of symmetry-allowed, switchable magnon spin transport in multiferroics, thereby establishing a critical framework for comprehending magnon transport within complex magnetic textures., Comment: 6 Pages, 3 Figures

Pattern formation in spin systems with continuous-rotational symmetry (CRS) provides a powerful platform to study emergent complex magnetic phases and topological defects in condensed-matter physics. However, its understanding and correlation with unconventional magnetic order along with high-resolution nanoscale imaging is challenging. Here, we employ scanning NV magnetometry to unveil the morphogenesis of spin cycloids at both the local and global scales within a single ferroelectric domain of (111)-oriented BiFeO$_3$ (which is a non-collinear antiferromagnet), resulting in formation of a glassy labyrinthine pattern. We find that the domains of locally oriented cycloids are interconnected by an array of topological defects and exhibit isotropic energy landscape predicted by first-principles calculations. We propose that the CRS of spin-cycloid propagation directions within the (111) drives the formation of the labyrinthine pattern and the associated topological defects such as antiferromagnetic skyrmions. Unexpectedly, reversing the as-grown ferroelectric polarization from [$\bar{1}$$\bar{1}$$\bar{1}$] to [111] induces a magnetic phase transition, destroying the labyrinthine pattern and producing a deterministic non-volatile non cycloidal, uniformly magnetized state. These findings highlight that (111)-oriented BiFeO$_3$ is not only important for studying the fascinating subject of pattern formation but could also be utilized as an ideal platform for integrating novel topological defects in the field of antiferromagnetic spintronics., Comment: 20 pages, 15 figures

As text-to-image models grow increasingly powerful and complex, their burgeoning size presents a significant obstacle to widespread adoption, especially on resource-constrained devices. This paper presents a pioneering study on post-training pruning of Stable Diffusion 2, addressing the critical need for model compression in text-to-image domain. Our study tackles the pruning techniques for the previously unexplored multi-modal generation models, and particularly examines the pruning impact on the textual component and the image generation component separately. We conduct a comprehensive comparison on pruning the model or the single component of the model in various sparsities. Our results yield previously undocumented findings. For example, contrary to established trends in language model pruning, we discover that simple magnitude pruning outperforms more advanced techniques in text-to-image context. Furthermore, our results show that Stable Diffusion 2 can be pruned to 38.5% sparsity with minimal quality loss, achieving a significant reduction in model size. We propose an optimal pruning configuration that prunes the text encoder to 47.5% and the diffusion generator to 35%. This configuration maintains image generation quality while substantially reducing computational requirements. In addition, our work uncovers intriguing questions about information encoding in text-to-image models: we observe that pruning beyond certain thresholds leads to sudden performance drops (unreadable images), suggesting that specific weights encode critical semantics information. This finding opens new avenues for future research in model compression, interoperability, and bias identification in text-to-image models. By providing crucial insights into the pruning behavior of text-to-image models, our study lays the groundwork for developing more efficient and accessible AI-driven image generation systems

Diffusive propagation of spin waves and their quanta - magnons - in the archetypal magnetic insulator yttrium iron garnet (YIG) is under a surge of research for low-power and low-loss data communication. However, operation under external magnetic fields reduces magnon diffusion length, attenuates the voltage amplitude at measurement terminals, and complicates the architecture of magnonic devices. Here, we explore the low-field and field-free regime of diffusive magnon transport in YIG films. We demonstrate that the field-induced suppression of magnon diffusion length can be fully inhibited only at the zero-field limit. Even a modest field of 10mT attenuates the non-local spin voltage by $\sim$ 20$\%$ in a transport channel of $\sim$ 1$\mu$m long. Using Stoner-Wohlfarth macrospin simulations, we reveal that an often overlooked, in-plane uniaxial anisotropy becomes the critical parameter governing the field-free operation of magnonic devices. We further demonstrate a tenfold enhancement in the effective field associated with the in-plane uniaxial anisotropy of YIG films at low temperatures - a key finding for field-free operation of magnonic devices under cryogenic conditions., Comment: 25 pages, 5 figures

Auto-SPICE is the first fully automated framework leveraging large language models (LLMs) to generate Simulation Programs with Integrated Circuit Emphasis (SPICE) netlists. It addresses a long-standing challenge in automating netlist generation for analog circuits within circuit design automation. Automating this workflow could accelerate the creation of finetuned LLMs for analog circuit design and verification. We identify key challenges in this automation and evaluate the multi-modal capabilities of state-of-the-art LLMs, particularly GPT-4, to address these issues. We propose a three-step workflow to overcome current limitations: labeling analog circuits, prompt tuning, and netlist verification. This approach aims to create an end-to-end SPICE netlist generator from circuit schematic images, tackling the long-standing hurdle of accurate netlist generation. Our framework demonstrates significant performance improvements, tested on approximately 2,100 schematics of varying complexity. We open-source this solution for community-driven development.

We present a novel hybrid quantum-classical vision transformer architecture incorporating quantum orthogonal neural networks (QONNs) to enhance performance and computational efficiency in high-energy physics applications. Building on advancements in quantum vision transformers, our approach addresses limitations of prior models by leveraging the inherent advantages of QONNs, including stability and efficient parameterization in high-dimensional spaces. We evaluate the proposed architecture using multi-detector jet images from CMS Open Data, focusing on the task of distinguishing quark-initiated from gluon-initiated jets. The results indicate that embedding quantum orthogonal transformations within the attention mechanism can provide robust performance while offering promising scalability for machine learning challenges associated with the upcoming High Luminosity Large Hadron Collider. This work highlights the potential of quantum-enhanced models to address the computational demands of next-generation particle physics experiments., Comment: 9 pages, 7 figures

Accurate modeling of lithium ion (li-ion) batteries is essential for enhancing the safety, and efficiency of electric vehicles and renewable energy systems. This paper presents a data-inspired approach for improving the fidelity of reduced-order li-ion battery models. The proposed method combines a Genetic Algorithm with Sequentially Thresholded Ridge Regression (GA-STRidge) to identify and compensate for discrepancies between a low-fidelity model (LFM) and data generated either from testing or a high-fidelity model (HFM). The hybrid model, combining physics-based and data-driven methods, is tested across different driving cycles to demonstrate the ability to significantly reduce the voltage prediction error compared to the baseline LFM, while preserving computational efficiency. The model robustness is also evaluated under various operating conditions, showing low prediction errors and high Pearson correlation coefficients for terminal voltage in unseen environments., Comment: 6 pages

This evaluate the performance of geospatial image processing using two distinct data storage formats: Zarr and TIFF. Geospatial images, converted to numerous applications like environmental monitoring, urban planning, and disaster management. Traditional Tagged Image File Format is mostly used because it is simple and compatible but may lack by performance limitations while working on large datasets. Zarr is a new format designed for the cloud systems,that offers scalability and efficient storage with data chunking and compression techniques. This study compares the two formats in terms of storage efficiency, access speed, and computational performance during typical geospatial processing tasks. Through analysis on a range of geospatial datasets, this provides details about the practical advantages and limitations of each format,helping users to select the appropriate format based on their specific needs and constraints.

Interactive sensors are an important component of robotic systems but often require manual replacement due to wear and tear. Automating this process can enhance system autonomy and facilitate long-term deployment. We developed an autonomous sensor exchange and calibration system for an agriculture crop monitoring robot that inserts a nitrate sensor into cornstalks. A novel gripper and replacement mechanism, featuring a reliable funneling design, were developed to enable efficient and reliable sensor exchanges. To maintain consistent nitrate sensor measurement, an on-board sensor calibration station was integrated to provide in-field sensor cleaning and calibration. The system was deployed at the Ames Curtis Farm in June 2024, where it successfully inserted nitrate sensors with high accuracy into 30 cornstalks with a 77$\%$ success rate.

We report on the onset of a coronal mass ejection (CME) using spectroscopic observations in 5303 {\AA} coronal emission line with the Visible Emission Line Coronagraph (VELC) onboard ADITYA-L1, the recently launched first Indian space solar mission. The CME was observed on 16 July 2024 in association with a X1.9 class soft X-ray flare from heliographic location S05W85. The VELC observations were near the west limb of Sun during the CME. The results obtained helped to constrain the onset time of the CME. In addition, they indicate ${\approx}$50% decrease in the coronal intensity near the source region of the CME due to mass depletion, ${\approx}$15% enhancement in the emission line width, and redshifted Doppler velocity of about ${\approx}10$ km/s. The non-thermal velocity associated with the line broadening is ${\approx}24.87$ km/s.

Advances in quantum algorithms suggest a tentative scaling advantage on certain combinatorial optimization problems. Recent work, however, has also reinforced the idea that barren plateaus render variational algorithms ineffective on large Hilbert spaces. Hence, finding annealing protocols by variation ultimately appears to be difficult. Similarly, the adiabatic theorem fails on hard problem instances with first-order quantum phase transitions. Here, we show how to use the spin coherent-state path integral to shape the geometry of quantum adiabatic evolution, leading to annealing protocols at polynomial overhead that provide orders-of-magnitude improvements in the probability to measure optimal solutions, relative to linear protocols. These improvements are not obtained on a controllable toy problem but on randomly generated hard instances (Sherrington-Kirkpatrick and Maximum 2-Satisfiability), making them generic and robust. Our method works for large systems and may thus be used to improve the performance of state-of-the-art quantum devices., Comment: 12 pages, 11 figures

We investigate how magnetic field variations around accreting black holes on event horizon scales affect the morphology of magnetically-driven jet on larger scales. By performing radiative transfer calculations on general relativistic magnetohydrodynamics simulations, we find that temporal variation in the magnetic flux on the event horizon and the jet power are imprinted on the variability of jet width up to several hundred gravitational radii. When the magnetic flux around the black hole drops and then rises, the jet initially narrows or becomes truncated, then widens, creating a thin-thick pattern that propagates down the jet. This suggests that extended jet observations can provide a history record of horizon-scale magnetic field dynamics, and conversely, upcoming changes in the jet image can be predicted from direct observation of the magnetized accreting plasma near the black hole. Furthermore, the pattern of jet width variations shows acceleration up to the relativistic regime as it moves away from the black hole, aligning with plasma bulk motion. We also find in time-averaged images that both the bulk plasma motion and magnetic field configuration in the jet-launching region, which are sensitive to black hole spin, shape diverse features through relativistic beaming and aberration. Higher black hole spins result in more poloidal bulk motion and toroidal magnetic fields, leading to more symmetric jet images and linear polarization patterns. These results suggest a new method for testing the magnetically arrested disk model and the Blandford-Znajek process, and for determining the black hole spin through observations bridging horizon and jet-launching scales., Comment: 19 pages, 16 figures, submitted to ApJ, movie available at https://youtu.be/FwiKspIXjZk

Compared to rigid hands, underactuated compliant hands offer greater adaptability to object shapes, provide stable grasps, and are often more cost-effective. However, they introduce uncertainties in hand-object interactions due to their inherent compliance and lack of precise finger proprioception as in rigid hands. These limitations become particularly significant when performing contact-rich tasks like insertion. To address these challenges, additional sensing modalities are required to enable robust insertion capabilities. This letter explores the essential sensing requirements for successful insertion tasks with compliant hands, focusing on the role of visuotactile perception. We propose a simulation-based multimodal policy learning framework that leverages all-around tactile sensing and an extrinsic depth camera. A transformer-based policy, trained through a teacher-student distillation process, is successfully transferred to a real-world robotic system without further training. Our results emphasize the crucial role of tactile sensing in conjunction with visual perception for accurate object-socket pose estimation, successful sim-to-real transfer and robust task execution.

It is quite intriguing to investigate the transition from a topological insulator (TI) phase to topological crystalline insulator (TCI) phase in a material as the latter has an advantage over the former in controlled device applications. This work investigates the existence of this dual topological behavior in Sn-based ternary chalcogenides family PbSnX2 (X=S, Se, Te) under the hydrostatic pressure using first-principles approach. These materials are dynamically sTABLE at ambient and elevated pressure conditions up to which the topological phase transitions (TPTs) are studied. This family have a topologically trivial ground state with direct band gap values 0.338 eV, 0.183 eV and 0.217 eV for PbSnS2, PbSnSe2 and PbSnTe2, respectively. The first TPT i.e., TI phase for these materials is observed, under the effect of external pressure of 5 GPa, 2.5 GPa and 3.5 GPa, with a single band inversion at F-point in the bulk band structure and an odd number of Dirac cones along the (111) surface. A further increase in pressure to 5.5 GPa, 3 GPa and 4 GPa results in another band inversion at {\Gamma}-point and an even number of Dirac cones along the (111) plane. These even number of band inversions suggest that (\(\bar{1}2\bar{1}\)) surface has mirror symmetry around (\(\bar{1}0\bar{1}\)) plane and hence, the TCI phase is obtained. This TCI phase is further corroborated with even value of mirror Chern number calculated using winding of Wannier charge centers., Comment: 21 pages, 6 figures, 2 tables

Silicon photonics is a leading platform for realizing the vast numbers of physical qubits needed for useful quantum information processing because it leverages mature complementary metal-oxide-semiconductor (CMOS) manufacturing to integrate on-chip thousands of optical devices for generating and manipulating quantum states of light. A challenge to the practical operation and scale-up of silicon quantum-photonic integrated circuits, however, is the need to control their extreme sensitivity to process and temperature variations, free-carrier and self-heating nonlinearities, and thermal crosstalk. To date these challenges have been partially addressed using bulky off-chip electronics, sacrificing many benefits of a chip-scale platform. Here, we demonstrate the first electronic-photonic quantum system-on-chip (EPQSoC) consisting of quantum-correlated photon-pair sources stabilized via on-chip feedback control circuits, all fabricated in a high-volume 45nm CMOS microelectronics foundry. We use non-invasive photocurrent sensing in a tunable microring cavity photon-pair source to actively lock it to a fixed pump laser while operating in the quantum regime, enabling large scale microring-based quantum systems. In this first demonstration of such a capability, we achieve a high CAR of 134 with an ultra-low g(2)(0) of 0.021 at 2.2 kHz off-chip detected pair rate and 3.3 MHz/mW2 on-chip pair generation efficiency, and over 100 kHz off-chip detected pair rate at higher pump powers (1.5 MHz on-chip). These sources maintain stable quantum properties in the presence of temperature variations, operating reliably in practical settings with many adjacent devices creating thermal disturbances on the same chip. Such dense electronic-photonic integration enables implementation and control of quantum-photonic systems at the scale required for useful quantum information processing with CMOS-fabricated chips.

We construct motivic cohomology cycles in the group $H^3_{\mathcal M}(Z,{\mathbb Q}(2))$ where $Z$ is a K3 surface obtained as a double cover of a del Pezzo surface $X$ branched at a curve in $|-2K_X|$. The construction uses (-1) curves on the del Pezzo and is a generalization of a recent pre-print of Ken Sato arXiv: 2408.09102 where he considers the case of fourfold covers of ${\mathbb P}^2$ branched at a quartic curve., Comment: 14 pages, 1 figure

In this study, we introduce a novel multi-modal biometric authentication system that integrates facial, vocal, and signature data to enhance security measures. Utilizing a combination of Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs), our model architecture uniquely incorporates dual shared layers alongside modality-specific enhancements for comprehensive feature extraction. The system undergoes rigorous training with a joint loss function, optimizing for accuracy across diverse biometric inputs. Feature-level fusion via Principal Component Analysis (PCA) and classification through Gradient Boosting Machines (GBM) further refine the authentication process. Our approach demonstrates significant improvements in authentication accuracy and robustness, paving the way for advanced secure identity verification solutions.

Ridesharing has experienced significant global growth over the past decade and is becoming integral to future transportation networks. These services offer alternative mobility options in many urban areas, promoting car-light or car-free lifestyles, with their market share rapidly expanding due to the convenience they offer. However, alongside these benefits, concerns have arisen about the environmental impact of ridesharing, particularly its contribution to carbon emissions. A major source of these emissions is deadhead miles that are driven without passengers between trips. This issue is especially pronounced during high-demand periods when the number of ride requests exceeds platform capacity, leading to longer deadhead miles and higher emissions. While reducing these unproductive miles can lower emissions, it may also result in longer wait times for passengers as they wait for a nearby driver, potentially diminishing the overall user experience. In this paper, we propose LARA, an online algorithm for rider-to-driver assignment that dynamically adjusts the maximum allowed deadhead miles for drivers and assigns ride requests accordingly. While LARA can be applied under any conditions, it is particularly more effective during high-demand hours, aiming to reduce both carbon emissions and rider wait times. We prove that LARA achieves near-optimal performance in online settings compared to the optimal offline algorithm. Furthermore, we evaluate LARA using both synthetic and real-world datasets, demonstrating up to 34.2% reduction in emissions and up to 42.9% reduction in rider wait times compared to state-of-the-art algorithms. While recent studies have introduced the problem of emission-aware ride assignment, LARA is the first algorithm to provide both theoretical and empirical guarantees on performance., Comment: 20 pages

One Health issues, such as the spread of highly pathogenic avian influenza (HPAI), present significant challenges at the intersection of human, animal, and environmental health. Recent H5N1 outbreaks underscore the need for comprehensive modeling that capture the complex interactions between various entities in these interconnected ecosystems, encompassing livestock, wild birds, and human populations. To support such efforts, we present a synthetic spatiotemporal gridded dataset for the contiguous United States, referred to as a digital similar. The methodology for constructing this digital similar involves fusing diverse datasets using statistical and optimization techniques. The livestock component includes farm-level representations of multiple livestock types -- cattle, poultry, hogs, and sheep -- including further categorization into subtypes, such as milk and beef cows, chicken, turkeys, ducks, etc. It also includes location-level data for livestock-product processing centers. Weekly abundance data for key wild bird species involved in avian flu transmission are included along with temporal networks of movements. Gridded distributions of the human population, along with demographic and occupational features, capture the placement of agricultural workers and the general population. The digital similar is verified and validated in multiple ways.This dataset aims to provide a comprehensive basis for modeling complex phenomena at the wild-domestic-human interfaces.

Traditionally, digital hardware designs are written in the Verilog hardware description language (HDL) and debugged manually by engineers. This can be time-consuming and error-prone for complex designs. Large Language Models (LLMs) are emerging as a potential tool to help generate fully functioning HDL code, but most works have focused on generation in the single-shot capacity: i.e., run and evaluate, a process that does not leverage debugging and as such does not adequately reflect a realistic development process. In this work we evaluate the ability of LLMs to leverage feedback from electronic design automation (EDA) tools to fix mistakes in their own generated Verilog. To accomplish this we present an open-source, highly customizable framework, AutoChip, which combines conversational LLMs with the output from Verilog compilers and simulations to iteratively generate and repair Verilog. To determine the success of these LLMs we leverage the VerilogEval benchmark set. We evaluate four state-of-the-art conversational LLMs, focusing on readily accessible commercial models. EDA tool feedback proved to be consistently more effective than zero-shot prompting only with GPT-4o, the most computationally complex model we evaluated. In the best case we observed a 5.8% increase in the number of successful designs with a 34.2% decrease in cost over the best zero-shot results. Mixing smaller models with this larger model at the end of the feedback iterations resulted in equally as much success as with GPT-4o using feedback, but for an additional 41.9% less cost (overall decrease in cost over zero-shot of 89.6%).

This work is an attempt to classify and quantify instances when a weighted sum of two squares of positive integers, $3n_{1}^2+n_{2}^2$, can be realized in more than one way. Our project was inspired by a particular study of two-dimensional quantum billiards [S. G. Jackson, H. Perrin, G. E. Astrakharchik, and M. Olshanii, SciPost Phys. Core 7, 062 (2024)] where the weighted sums of interest represents an energy level with the two integers being the billiard's quantum numbers; there, the 3-fold degeneracies seem to dominate the energy spectrum. Interestingly, contrary to the conventional paradigm, these degeneracies are not caused by some non-commuting symmetries of the system.

The increasing prevalence of AI-generated content alongside human-written text underscores the need for reliable discrimination methods. To address this challenge, we propose a novel framework with textual embeddings from Pre-trained Language Models (PLMs) to distinguish AI-generated and human-authored text. Our approach utilizes Embedding Fusion to integrate semantic information from multiple Language Models, harnessing their complementary strengths to enhance performance. Through extensive evaluation across publicly available diverse datasets, our proposed approach demonstrates strong performance, achieving classification accuracy greater than 96% and a Matthews Correlation Coefficient (MCC) greater than 0.93. This evaluation is conducted on a balanced dataset of texts generated from five well-known Large Language Models (LLMs), highlighting the effectiveness and robustness of our novel methodology.

In safeguarding mission-critical systems, such as Unmanned Aerial Vehicles (UAVs), preserving the privacy of path trajectories during navigation is paramount. While the combination of Reinforcement Learning (RL) and Fully Homomorphic Encryption (FHE) holds promise, the computational overhead of FHE presents a significant challenge. This paper proposes an innovative approach that leverages Knowledge Distillation to enhance the practicality of secure UAV navigation. By integrating RL and FHE, our framework addresses vulnerabilities to adversarial attacks while enabling real-time processing of encrypted UAV camera feeds, ensuring data security. To mitigate FHE's latency, Knowledge Distillation is employed to compress the network, resulting in an impressive 18x speedup without compromising performance, as evidenced by an R-squared score of 0.9499 compared to the original model's score of 0.9631. Our methodology underscores the feasibility of processing encrypted data for UAV navigation tasks, emphasizing security alongside performance efficiency and timely processing. These findings pave the way for deploying autonomous UAVs in sensitive environments, bolstering their resilience against potential security threats., Comment: arXiv admin note: text overlap with arXiv:2404.17225

In real-world applications, the distribution of the data, and our goals, evolve over time. The prevailing theoretical framework for studying machine learning, namely probably approximately correct (PAC) learning, largely ignores time. As a consequence, existing strategies to address the dynamic nature of data and goals exhibit poor real-world performance. This paper develops a theoretical framework called "Prospective Learning" that is tailored for situations when the optimal hypothesis changes over time. In PAC learning, empirical risk minimization (ERM) is known to be consistent. We develop a learner called Prospective ERM, which returns a sequence of predictors that make predictions on future data. We prove that the risk of prospective ERM converges to the Bayes risk under certain assumptions on the stochastic process generating the data. Prospective ERM, roughly speaking, incorporates time as an input in addition to the data. We show that standard ERM as done in PAC learning, without incorporating time, can result in failure to learn when distributions are dynamic. Numerical experiments illustrate that prospective ERM can learn synthetic and visual recognition problems constructed from MNIST and CIFAR-10., Comment: Accepted to NeurIPS 2024

Measurement-based quantum computing offers a promising route towards scalable, universal photonic quantum computation. This approach relies on the deterministic and efficient generation of photonic graph states in which many photons are mutually entangled with various topologies. Recently, deterministic sources of graph states have been demonstrated with quantum emitters in both the optical and microwave domains. In this work, we demonstrate deterministic and reconfigurable graph state generation with optical solid-state integrated quantum emitters. Specifically, we use a single semiconductor quantum dot in a cavity to generate caterpillar graph states, the most general type of graph state that can be produced with a single emitter. By using fast detuned optical pulses, we achieve full control over the spin state, enabling us to vary the entanglement topology at will. We perform quantum state tomography of two successive photons, measuring Bell state fidelities up to 0.80$\pm$0.04 and concurrences up to 0.69$\pm$0.09, while maintaining high photon indistinguishability. This simple optical scheme, compatible with commercially available quantum dot-based single photon sources, brings us a step closer to fault-tolerant quantum computing with spins and photons.

The single variant spin cycloid and associated antiferromagnetic order in multiferroic BiFeO3 can provide a direct and predictable magnetoelectric coupling to ferroelectric order for deterministic switching, and also a key to fundamental understanding of spin transport and magnon-based applications in the system. (111) oriented BiFeO3 supplies an easy magnetic plane for the spin cycloid, but despite previous efforts, achieving deterministic switching of the single spin cycloid over multiple cycles remains challenging due to the presence of multiple spin cycloid domains in the (111) plane. Here we show that anisotropic in-plane strain engineering can stabilize a single antiferromagnetic domain and provide robust, deterministic switching. We grow BiFeO3 on orthorhombic NdGaO3 (011)o [(111)pc] substrates, breaking the spin cycloid degeneracy and by imposing a uniaxial strain in the (111) plane. This stabilization is confirmed through direct imaging with scanning NV microscopy and non-resonant X-ray diffraction. Remarkably, we achieved deterministic and non-volatile 180{\deg} switching of ferroelectric and associated antiferromagnetic domains over 1,000 cycles, significantly outperforming existing approaches. Our findings underscore that anisotropic strain engineering opens up exciting possibilities for (111)pc monodomain BiFeO3 in potential magnetoelectric and emerging magnonic applications., Comment: 17 pages, 4 figures

GPT-4o is an autoregressive omni model that accepts as input any combination of text, audio, image, and video, and generates any combination of text, audio, and image outputs. It's trained end-to-end across text, vision, and audio, meaning all inputs and outputs are processed by the same neural network. GPT-4o can respond to audio inputs in as little as 232 milliseconds, with an average of 320 milliseconds, which is similar to human response time in conversation. It matches GPT-4 Turbo performance on text in English and code, with significant improvement on text in non-English languages, while also being much faster and 50\% cheaper in the API. GPT-4o is especially better at vision and audio understanding compared to existing models. In line with our commitment to building AI safely and consistent with our voluntary commitments to the White House, we are sharing the GPT-4o System Card, which includes our Preparedness Framework evaluations. In this System Card, we provide a detailed look at GPT-4o's capabilities, limitations, and safety evaluations across multiple categories, focusing on speech-to-speech while also evaluating text and image capabilities, and measures we've implemented to ensure the model is safe and aligned. We also include third-party assessments on dangerous capabilities, as well as discussion of potential societal impacts of GPT-4o's text and vision capabilities.

Recent model merging methods demonstrate that the parameters of fully-finetuned models specializing in distinct tasks can be combined into one model capable of solving all tasks without retraining. Yet, this success does not transfer well when merging LoRA finetuned models. We study this phenomenon and observe that the weights of LoRA finetuned models showcase a lower degree of alignment compared to their fully-finetuned counterparts. We hypothesize that improving this alignment is key to obtaining better LoRA model merges, and propose KnOTS to address this problem. KnOTS uses the SVD to jointly transform the weights of different LoRA models into an aligned space, where existing merging methods can be applied. In addition, we introduce a new benchmark that explicitly evaluates whether merged models are general models. Notably, KnOTS consistently improves LoRA merging by up to 4.3% across several vision and language benchmarks, including our new setting. We release our code at: https://github.com/gstoica27/KnOTS.

Knowledge Editing (KE) algorithms alter models' internal weights to perform targeted updates to incorrect, outdated, or otherwise unwanted factual associations. In order to better define the possibilities and limitations of these approaches, recent work has shown that applying KE can adversely affect models' factual recall accuracy and diminish their general reasoning abilities. While these studies give broad insights into the potential harms of KE algorithms, e.g., via performance evaluations on benchmarks, we argue little is understood as to why such destructive failures occur. Is it possible KE methods distort representations of concepts beyond the targeted fact, hence hampering abilities at broad? If so, what is the extent of this distortion? To take a step towards addressing such questions, we define a novel synthetic task wherein a Transformer is trained from scratch to internalize a ``structured'' knowledge graph. The structure enforces relationships between entities of the graph, such that editing a factual association has "trickling effects" on other entities in the graph (e.g., altering X's parent is Y to Z affects who X's siblings' parent is). Through evaluations of edited models and analysis of extracted representations, we show that KE inadvertently affects representations of entities beyond the targeted one, distorting relevant structures that allow a model to infer unseen knowledge about an entity. We call this phenomenon representation shattering and demonstrate that it results in degradation of factual recall and reasoning performance more broadly. To corroborate our findings in a more naturalistic setup, we perform preliminary experiments with a pretrained GPT-2-XL model and reproduce the representation shattering effect therein as well. Overall, our work yields a precise mechanistic hypothesis to explain why KE has adverse effects on model capabilities., Comment: Under review

We consider a retailer running a switchback experiment for the price of a single product, with infinite supply. In each period, the seller chooses a price $p$ from a set of predefined prices that consist of a reference price and a few discounted price levels. The goal is to estimate the demand gradient at the reference price point, with the goal of adjusting the reference price to improve revenue after the experiment. In our model, in each period, a unit mass of buyers arrives on the market, with values distributed based on a time-varying process. Crucially, buyers are forward looking with a discounted utility and will choose to not purchase now if they expect to face a discounted price in the near future. We show that forward-looking demand introduces bias in naive estimators of the demand gradient, due to intertemporal interference. Furthermore, we prove that there is no estimator that uses data from price experiments with only two price points that can recover the correct demand gradient, even in the limit of an infinitely long experiment with an infinitesimal price discount. Moreover, we characterize the form of the bias of naive estimators. Finally, we show that with a simple three price level experiment, the seller can remove the bias due to strategic forward-looking behavior and construct an estimator for the demand gradient that asymptotically recovers the truth.

Battery geometry (shape and size) is one of the important parameters which governs the battery capacity and thermal behavior. In the dynamic conditions or during the operation, the performance of batteries become much more complex. Herein, the changes in thermal behavior of lithium-ion battery (LIB)by altering the geometry i.e., length to diameter ratio (l/d), is investigated. The geometries considered are named as large geometry (LG), datum geometry (DG) and small geometry (SG) with the l/d ratio of 5.25, 3.61, and 2.38, respectively. A three-dimensional (3D) multi-partition thermal model is adopted, and the numerical results are validated by the published experimental data. For three different cooling approaches such as radial, both-tab and mixed cooling, the average battery temperature and temperature heterogeneity are thoroughly examined considering the heat transfer coefficients (h) of50 and 100 W/m2K at discharge rates of 1, 2 and 3C. Amongst, the minimum average battery temperature is exhibited by DG, the minimum radial temperature heterogeneity is obtained from LG, and substantial outperformance in terms of faster cooling rate is identified for SG, irrespective of the cooling approach employed, Comment: 39 pages, 12 Figures

Lead-free metal halide perovskites are emerging as less-toxic alternatives to their lead-based counterparts. However, their applicability in optoelectronic devices is limited, and the charge transport dynamics remain poorly understood. Understanding photo-induced charge and structural dynamics is critical for unlocking the potential of these novel systems. In this work, we employ ultrafast optical and Raman spectroscopy combined with band structure calculations to investigate the coupled electronic and vibrational dynamics in Caesium gold bromide, a promising lead-free perovskite. We find that the band-edge charge transfer states are strongly coupled to Au-Br stretching phonon modes, leading to frequency modulation of absorption by impulsively excited coherent phonons. Early-stage relaxation is characterized by dynamics of delocalized charge transfer excitation and slowly decaying coherent phonons. The electronic and vibrational relaxation reveals a slow formation of a localized polaronic state in the 10-20 ps timescale. Using a displaced harmonic oscillator model, the polaronic binding energy is estimated to be ~80 meV following lattice relaxation along the phonon modes. Strong exciton-phonon coupling and slow polaron formation via coupling to lattice modes make this material a promising testbed for the control of coherent phonons and localized polaronic states using light., Comment: 38 pages, 15 Figures

Large Language Models (LLMs) have demonstrated remarkable performance across various domains, including healthcare. However, their ability to effectively represent structured non-textual data, such as the alphanumeric medical codes used in records like ICD-10 or SNOMED-CT, is limited and has been particularly exposed in recent research. This paper examines the challenges LLMs face in processing medical codes due to the shortcomings of current tokenization methods. As a result, we introduce the UniStruct architecture to design a multimodal medical foundation model of unstructured text and structured data, which addresses these challenges by adapting subword tokenization techniques specifically for the structured medical codes. Our approach is validated through model pre-training on both an extensive internal medical database and a public repository of structured medical records. Trained on over 1 billion tokens on the internal medical database, the proposed model achieves up to a 23% improvement in evaluation metrics, with around 2% gain attributed to our proposed tokenization. Additionally, when evaluated on the EHRSHOT public benchmark with a 1/1000 fraction of the pre-training data, the UniStruct model improves performance on over 42% of the downstream tasks. Our approach not only enhances the representation and generalization capabilities of patient-centric models but also bridges a critical gap in representation learning models' ability to handle complex structured medical data, alongside unstructured text., Comment: NeurIPS 2024 Workshop on Unifying Representations in Neural Models (UniReps 2024)

Coupling black hole (BH) feeding and feedback involves interactions across vast spatial and temporal scales that is computationally challenging. Tracking gas inflows and outflows from kilo-parsec scales to the event horizon for non-spinning BHs in the presence of strong magnetic fields, Cho et al. (2023, 2024) report strong suppression of accretion on horizon scales and low (2%) feedback efficiency. In this letter, we explore the impact of these findings for the supermassive BHs M87* and Sgr A*, using high-resolution, non-cosmological, magnetohydrodynamic (MHD) simulations with the Feedback In Realistic Environments (FIRE-2) model. With no feedback, we find rapid BH growth due to "cooling flows," and for 2% efficiency feedback, while accretion is suppressed, the rates still remain higher than constraints from Event Horizon Telescope (EHT) data (Event Horizon Telescope Collaboration et al. 2021, 2022) for M87* and Sgr A*. To match EHT observations of M87*, a feedback efficiency greater than 15% is required, suggesting the need to include enhanced feedback from BH spin. Similarly, a feedback efficiency of $>15\%$ is needed for Sgr A* to match the estimated observed star formation rate of $\lesssim 2 {\rm M_\odot}$ yr$^{-1}$. However, even with 100% feedback efficiency, the accretion rate onto Sgr A* matches with EHT data only on rare occasions in the simulations, suggesting that Sgr A* is likely in a temporary quiescent phase currently. Bridging accretion and feedback across scales, we conclude that higher feedback efficiency, possibly due to non-zero BH spin, is necessary to suppress "cooling flows" and match observed accretion and star formation rates in M87* and Sgr A*., Comment: 13 pages, 4 figures

This paper proposes a novel metamaterial-inspired absorber designed to enhance the data rate in 52% RF to DC conversion simultaneous wireless information and power transfer system (SWIPT) through biological tissue. The proposed absorber includes split-ring resonators(SRRs) and demonstrates significant permeability characteristics, with both the real and imaginary parts being negative and close to -1. It also improves isolation by around 5dB in a WPT distance of 9mm. A 5mm thick phantom is used for biological tissue in this study. Experimental results exhibits that the SWIPT systems including a rectifier that converts 52% RF to DC efficiency in a WPT distance of 9mm embedding this absorber between power and signal ports at Tx side results in a 5dB improvement in isolation performance. By using proposed absorber, it enables a 7MB/s improvement of data rate and allows signals to be transmitted with 5dBm weaker power than without absorber SWIPT system.

Despite efforts from cloud and content providers to lower latency to acceptable levels for current and future services (e.g., augmented reality or cloud gaming), there are still opportunities for improvement. A major reason that traffic engineering efforts are challenged to lower latency is that the Internet's inter-domain routing protocol, the Border Gateway Protocol, is oblivious to any performance metric, and circuitous routing is still pervasive. In this work, we propose two implementation modifications that networks can leverage to make BGP latency-aware and reduce excessive latency inflation. These proposals, latency-proportional AS prepending and local preference neutralization, show promise towards providing a method for propagating abstract latency information with a reasonable increase in routing overhead.

Deep Neural Networks (DNNs) are increasingly deployed across applications. However, ensuring their reliability remains a challenge, and in many situations, alternative models with similar functionality and accuracy are available. Traditional accuracy-based evaluations often fail to capture behavioral differences between models, especially with limited test datasets, making it difficult to select or combine models effectively. Differential testing addresses this by generating test inputs that expose discrepancies in DNN model behavior. However, existing approaches face significant limitations: many rely on model internals or are constrained by available seed inputs. To address these challenges, we propose DiffGAN, a black-box test image generation approach for differential testing of DNN models. DiffGAN leverages a Generative Adversarial Network (GAN) and the Non-dominated Sorting Genetic Algorithm II to generate diverse and valid triggering inputs that reveal behavioral discrepancies between models. DiffGAN employs two custom fitness functions, focusing on diversity and divergence, to guide the exploration of the GAN input space and identify discrepancies between models' outputs. By strategically searching this space, DiffGAN generates inputs with specific features that trigger differences in model behavior. DiffGAN is black-box, making it applicable in more situations. We evaluate DiffGAN on eight DNN model pairs trained on widely used image datasets. Our results show DiffGAN significantly outperforms a SOTA baseline, generating four times more triggering inputs, with greater diversity and validity, within the same budget. Additionally, the generated inputs improve the accuracy of a machine learning-based model selection mechanism, which selects the best-performing model based on input characteristics and can serve as a smart output voting mechanism when using alternative models.

Rapid adoption of machine learning (ML) technologies has led to a surge in power consumption across diverse systems, from tiny IoT devices to massive datacenter clusters. Benchmarking the energy efficiency of these systems is crucial for optimization, but presents novel challenges due to the variety of hardware platforms, workload characteristics, and system-level interactions. This paper introduces MLPerf Power, a comprehensive benchmarking methodology with capabilities to evaluate the energy efficiency of ML systems at power levels ranging from microwatts to megawatts. Developed by a consortium of industry professionals from more than 20 organizations, MLPerf Power establishes rules and best practices to ensure comparability across diverse architectures. We use representative workloads from the MLPerf benchmark suite to collect 1,841 reproducible measurements from 60 systems across the entire range of ML deployment scales. Our analysis reveals trade-offs between performance, complexity, and energy efficiency across this wide range of systems, providing actionable insights for designing optimized ML solutions from the smallest edge devices to the largest cloud infrastructures. This work emphasizes the importance of energy efficiency as a key metric in the evaluation and comparison of the ML system, laying the foundation for future research in this critical area. We discuss the implications for developing sustainable AI solutions and standardizing energy efficiency benchmarking for ML systems., Comment: 14 pages, 11 figures, 1 table

Recent works show we can linearize large language models (LLMs) -- swapping the quadratic attentions of popular Transformer-based LLMs with subquadratic analogs, such as linear attention -- avoiding the expensive pretraining costs. However, linearizing LLMs often significantly degrades model quality, still requires training over billions of tokens, and remains limited to smaller 1.3B to 7B LLMs. We thus propose Low-rank Linear Conversion via Attention Transfer (LoLCATs), a simple two-step method that improves LLM linearizing quality with orders of magnitudes less memory and compute. We base these steps on two findings. First, we can replace an LLM's softmax attentions with closely-approximating linear attentions, simply by training the linear attentions to match their softmax counterparts with an output MSE loss ("attention transfer"). Then, this enables adjusting for approximation errors and recovering LLM quality simply with low-rank adaptation (LoRA). LoLCATs significantly improves linearizing quality, training efficiency, and scalability. We significantly reduce the linearizing quality gap and produce state-of-the-art subquadratic LLMs from Llama 3 8B and Mistral 7B v0.1, leading to 20+ points of improvement on 5-shot MMLU. Furthermore, LoLCATs does so with only 0.2% of past methods' model parameters and 0.4% of their training tokens. Finally, we apply LoLCATs to create the first linearized 70B and 405B LLMs (50x larger than prior work). When compared with prior approaches under the same compute budgets, LoLCATs significantly improves linearizing quality, closing the gap between linearized and original Llama 3.1 70B and 405B LLMs by 77.8% and 78.1% on 5-shot MMLU., Comment: 47 pages, 20 figures, 18 tables, preprint

Graph-based anomaly detection is pivotal in diverse security applications, such as fraud detection in transaction networks and intrusion detection for network traffic. Standard approaches, including Graph Neural Networks (GNNs), often struggle to generalize across shifting data distributions. Meanwhile, real-world domain knowledge is more stable and a common existing component of real-world detection strategies. To explicitly integrate such knowledge into data-driven models such as GCNs, we propose KnowGraph, which integrates domain knowledge with data-driven learning for enhanced graph-based anomaly detection. KnowGraph comprises two principal components: (1) a statistical learning component that utilizes a main model for the overarching detection task, augmented by multiple specialized knowledge models that predict domain-specific semantic entities; (2) a reasoning component that employs probabilistic graphical models to execute logical inferences based on model outputs, encoding domain knowledge through weighted first-order logic formulas. Extensive experiments on these large-scale real-world datasets show that KnowGraph consistently outperforms state-of-the-art baselines in both transductive and inductive settings, achieving substantial gains in average precision when generalizing to completely unseen test graphs. Further ablation studies demonstrate the effectiveness of the proposed reasoning component in improving detection performance, especially under extreme class imbalance. These results highlight the potential of integrating domain knowledge into data-driven models for high-stakes, graph-based security applications., Comment: Accepted to ACM CCS 2024

The difficulty of anonymizing text data hinders the development and deployment of NLP in high-stakes domains that involve private data, such as healthcare and social services. Poorly anonymized sensitive data cannot be easily shared with annotators or external researchers, nor can it be used to train public models. In this work, we explore the feasibility of using synthetic data generated from differentially private language models in place of real data to facilitate the development of NLP in these domains without compromising privacy. In contrast to prior work, we generate synthetic data for real high-stakes domains, and we propose and conduct use-inspired evaluations to assess data quality. Our results show that prior simplistic evaluations have failed to highlight utility, privacy, and fairness issues in the synthetic data. Overall, our work underscores the need for further improvements to synthetic data generation for it to be a viable way to enable privacy-preserving data sharing., Comment: Accepted to EMNLP 2024 (Findings)

This paper investigates decision-making in A/B experiments for online platforms and marketplaces. In such settings, due to constraints on inventory, A/B experiments typically lead to biased estimators because of interference; this phenomenon has been well studied in recent literature. By contrast, there has been relatively little discussion of the impact of interference on decision-making. In this paper, we analyze a benchmark Markovian model of an inventory-constrained platform, where arriving customers book listings that are limited in supply; our analysis builds on a self-contained analysis of general A/B experiments for Markov chains. We focus on the commonly used frequentist hypothesis testing approach for making launch decisions based on data from customer-randomized experiments, and we study the impact of interference on (1) false positive probability and (2) statistical power. We obtain three main findings. First, we show that for monotone treatments -- i.e., those where the treatment changes booking probabilities in the same direction relative to control in all states -- the false positive probability of the na\"ive difference-in-means estimator with classical variance estimation is correctly controlled. This result stems from a novel analysis of A/A experiments with arbitrary dependence structures, which may be of independent interest. Second, we demonstrate that for monotone treatments, the statistical power of this na\"ive approach is higher than that of any similar pipeline using a debiased estimator. Taken together, these two findings suggest that platforms may be better off not debiasing when treatments are monotone. Finally, using simulations, we investigate false positive probability and statistical power when treatments are non-monotone, and we show that the performance of the na\"ive approach can be arbitrarily worse than a debiased approach in such cases.

Using the semi-analytic model Serotina, we investigate the cosmic spin evolution of supermassive black holes incorporating recent results from general relativistic magnetohydrodynamics simulations of spin-down from relativistic jets. We compare several variations of our model with compiled black hole spin measurements derived from X-ray reflection spectroscopy, correcting for a bias arising from the spin-dependent radiative efficiency of accretion flows. We show that the observed spin distribution is in agreement with a model that includes jet-driven spin-down, a key mechanism that acts to modulate spins across cosmic time at both high and very low specific accretion rates. The data also clearly prefer models with coherent accretion over models in which accretion disks rapidly switch from prograde to retrograde. We further predict spin distributions accessible via spatially resolved event horizons by the next-generation Event Horizon Telescope (ngEHT) and Black Hole Explorer (BHEX), as well as gravitational waves by the Laser Interferometer Space Antenna (LISA), each of which offer unique and distinct windows into the population of spinning black holes. Jet-driven spin-down is most strongly imprinted on the abundance of very highly spinning objects in our model. In addition, we show that the spin distribution sampled by LISA events may contain a signature of the natal spin distribution of heavy seeds, but not of light seeds, offering additional discrimination between these seeding pathways. Spin distributions from these future observed samples can be used to constrain the detailed physical properties of the accretion flow on horizon scales around supermassive black holes., Comment: 23 pages, 11 figures, 4 tables. Submitted to ApJ