Search

Multimodal AI has the potential to significantly enhance document-understanding tasks, such as processing receipts, understanding workflows, extracting data from documents, and summarizing reports. Code generation tasks that require long-structured outputs can also be enhanced by multimodality. Despite this, their use in commercial applications is often limited due to limited access to training data and restrictive licensing, which hinders open access. To address these limitations, we introduce BigDocs-7.5M, a high-quality, open-access dataset comprising 7.5 million multimodal documents across 30 tasks. We use an efficient data curation process to ensure our data is high-quality and license-permissive. Our process emphasizes accountability, responsibility, and transparency through filtering rules, traceable metadata, and careful content analysis. Additionally, we introduce BigDocs-Bench, a benchmark suite with 10 novel tasks where we create datasets that reflect real-world use cases involving reasoning over Graphical User Interfaces (GUI) and code generation from images. Our experiments show that training with BigDocs-Bench improves average performance up to 25.8% over closed-source GPT-4o in document reasoning and structured output tasks such as Screenshot2HTML or Image2Latex generation. Finally, human evaluations showed a preference for outputs from models trained on BigDocs over GPT-4o. This suggests that BigDocs can help both academics and the open-source community utilize and improve AI tools to enhance multimodal capabilities and document reasoning. The project is hosted at https://bigdocs.github.io ., Comment: The project is hosted at https://bigdocs.github.io

Achieving quantum advantage in energy storage and power extraction is a primary objective in the design of quantum-based batteries. We explore how long-range (LR) interactions in conjunction with Floquet driving can improve the performance of quantum batteries, particularly when the battery is initialized in a fully polarized state. In particular, we exhibit that by optimizing the driving frequency, the maximum average power scales super extensively with system-size which is not achievable through next-nearest neighbor interactions or traditional unitary charging, thereby gaining genuine quantum advantage. We illustrate that the inclusion of either two-body or many-body interaction terms in the LR charging Hamiltonian leads to a scaling benefit. Furthermore, we discover that a super-linear scaling in power results from increasing the strength of interaction compared to the transverse magnetic field and the range of interaction with low fall-off rate, highlighting the advantageous role of long-range interactions in optimizing quantum battery charging., Comment: 10 pages, 8 figures

Proposals for quantum random access memory (QRAM) generally have a binary-tree structure, and thus require hardware that is exponential in the depth of the QRAM. For solid-state based devices, a fabrication yield that is less than $100\%$ implies that certain addresses at the bottom of the tree become inaccessible if a router in the unique path to that address is faulty. We discuss how to recover a functioning QRAM in the presence of faulty routers. We present the \texttt{IterativeRepair} algorithm, which constructs QRAMs layer by layer until the desired depth is reached. This algorithm utilizes ancilla flag qubits which reroute queries to faulty routers. We present a classical algorithm \texttt{FlagQubitMinimization} that attempts to minimize the required number of such ancilla. For a router failure rate of $1\%$ and a QRAM of depth $n=13$, we expect that on average 430 addresses need repair: we require only 1.5 ancilla flag qubits on average to perform this rerouting., Comment: 13 pages, 11 figures, associated code available https://github.com/dkweiss31/QRAMfaultyrouters

This white paper, developed through close collaboration between IBM Research and UIUC researchers within the IIDAI Institute, envisions transforming hybrid cloud systems to meet the growing complexity of AI workloads through innovative, full-stack co-design approaches, emphasizing usability, manageability, affordability, adaptability, efficiency, and scalability. By integrating cutting-edge technologies such as generative and agentic AI, cross-layer automation and optimization, unified control plane, and composable and adaptive system architecture, the proposed framework addresses critical challenges in energy efficiency, performance, and cost-effectiveness. Incorporating quantum computing as it matures will enable quantum-accelerated simulations for materials science, climate modeling, and other high-impact domains. Collaborative efforts between academia and industry are central to this vision, driving advancements in foundation models for material design and climate solutions, scalable multimodal data processing, and enhanced physics-based AI emulators for applications like weather forecasting and carbon sequestration. Research priorities include advancing AI agentic systems, LLM as an Abstraction (LLMaaA), AI model optimization and unified abstractions across heterogeneous infrastructure, end-to-end edge-cloud transformation, efficient programming model, middleware and platform, secure infrastructure, application-adaptive cloud systems, and new quantum-classical collaborative workflows. These ideas and solutions encompass both theoretical and practical research questions, requiring coordinated input and support from the research community. This joint initiative aims to establish hybrid clouds as secure, efficient, and sustainable platforms, fostering breakthroughs in AI-driven applications and scientific discovery across academia, industry, and society., Comment: 70 pages, 27 figures

Modern applications demand high performance and cost efficient database management systems (DBMSs). Their workloads may be diverse, ranging from online transaction processing to analytics and decision support. The cloud infrastructure enables disaggregation of monolithic DBMSs into components that facilitate software-hardware co-design. This is realized using pools of hardware resources, i.e., CPUs, GPUs, memory, FPGA, NVM, etc., connected using high-speed networks. This disaggregation trend is being adopted by cloud DBMSs because hardware re-provisioning can be achieved by simply invoking software APIs. Disaggregated DBMSs separate processing from storage, enabling each to scale elastically and independently. They may disaggregate compute usage based on functionality, e.g., compute needed for writes from compute needed for queries and compute needed for compaction. They may also use disaggregated memory, e.g., for intermediate results in a shuffle or for remote caching. The DBMS monitors the characteristics of a workload and dynamically assembles its components that are most efficient and cost effective for the workload. This paper is a summary of a panel session that discussed the capability, challenges, and opportunities of these emerging DBMSs and disaggregated hardware systems., Comment: This paper appeared in the {\em Performance Evaluation and Benchmarking} - 14th TPC Technology Conference, TPCTC 2022, Sydney, NSW, Australia, September 5, 2022, Revised Selected Papers. Lecture Notes in Computer Science 13860, Springer 2023, ISBN 978-3-031-29575-1

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.

This research paper presents the development of an AI model utilizing YOLOv8 for real-time weapon detection, aimed at enhancing safety in public spaces such as schools, airports, and public transportation systems. As incidents of violence continue to rise globally, there is an urgent need for effective surveillance technologies that can quickly identify potential threats. Our approach focuses on leveraging advanced deep learning techniques to create a highly accurate and efficient system capable of detecting weapons in real-time video streams. The model was trained on a comprehensive dataset containing thousands of images depicting various types of firearms and edged weapons, ensuring a robust learning process. We evaluated the model's performance using key metrics such as precision, recall, F1-score, and mean Average Precision (mAP) across multiple Intersection over Union (IoU) thresholds, revealing a significant capability to differentiate between weapon and non-weapon classes with minimal error. Furthermore, we assessed the system's operational efficiency, demonstrating that it can process frames at high speeds suitable for real-time applications. The findings indicate that our YOLOv8-based weapon detection model not only contributes to the existing body of knowledge in computer vision but also addresses critical societal needs for improved safety measures in vulnerable environments. By harnessing the power of artificial intelligence, this research lays the groundwork for developing practical solutions that can be deployed in security settings, ultimately enhancing the protective capabilities of law enforcement and public safety agencies., Comment: 21 pages, 5 figures

High-fidelity quantum optomechanical control of a mechanical oscillator requires the ability to perform efficient, low-noise operations on long-lived phononic excitations. Microfabricated high-overtone bulk acoustic wave resonators ($\mathrm{\mu}$HBARs) have been shown to support high-frequency (> 10 GHz) mechanical modes with exceptionally long coherence times (> 1.5 ms), making them a compelling resource for quantum optomechanical experiments. In this paper, we demonstrate a new optomechanical system that permits quantum optomechanical control of individual high-coherence phonon modes supported by such $\mathrm{\mu}$HBARs for the first time. We use this system to perform laser cooling of such ultra-massive (7.5 $\mathrm{\mu}$g) high frequency (12.6 GHz) phonon modes from an occupation of ${\sim}$22 to fewer than 0.4 phonons, corresponding to laser-based ground-state cooling of the most massive mechanical object to date. Through these laser cooling experiments, no absorption-induced heating is observed, demonstrating the resilience of the $\mathrm{\mu}$HBAR against parasitic heating. The unique features of such $\mathrm{\mu}$HBARs make them promising as the basis for a new class of quantum optomechanical systems that offer enhanced robustness to decoherence, necessary for efficient, low-noise photon-phonon conversion.

In this article, we present our recent experiments on fission fragment rocket propulsion, and on an innovative new design for an alpha particle detection system that has been inspired by these rocketry results. Our test platform, which operates within high magnetic fields 3 T over a large cross$\unicode{x2013}$section (approximately 40 cm in diameter), has been used as a test platform to evaluate the containment and thrust within a future fission$\unicode{x2013}$fragment rocket engine (FFRE). This much more efficient nuclear rocket propulsion FFRE design was first proposed in the 1980s with the intent of greatly reducing transit times in long$\unicode{x2013}$duration space travel. Our objective is to enhance the operational efficiency of this nuclear rocket while gaining deeper insights into the behavior of fuel particles and of the fission$\unicode{x2013}$fragment ejecta within strong magnetic fields experimentally. Through a combination of simulations and experimental work, we established a method for the production and detection of alpha particles as a surrogate for fission fragments. To achieve this, we employed Americium$\unicode{x2013}$241 ($^{ 241}$Am) sources, which were situated within a cylindrical vacuum chamber positioned in a 3$\unicode{x2013}$T Siemens MRI superconducting magnet. By simulating, measuring, and analyzing the emitted alpha particle flux, we gained valuable information about the distribution and likelihood of escape of fission fragments in a future FFRE design. This approach could potentially achieve both high specific impulse and power density in advanced nuclear propulsion systems, such as the FFRE. More generally, this work provides a powerful new approach for analyzing ion flux and nuclear particle or nuclear reaction fragments from a wide variety of experimental designs., Comment: 16 pages, 6 figures, and one data table

We present strain tuning of excitonic emission in monolayer MoSe$_2$ by using a high-temperature physical vapor deposition (PVD). The use of two amorphous substrates, Si$_{3}$N$_{4}$ and SiO$_{2}$, provides two setpoints to induce distinct amounts of \textit{biaxial} tensile strain determined by a thermal expansion mismatch between the monolayer and the substrate. The tuning rate of the $A$-exciton transition energy is found to be 103 meV/\% by photoluminescence (PL), which represents the highest value realized by biaxial strain in transition metal dichalcogenides. The biaxial nature of the tensile strain is confirmed by polarization-resolved second harmonic generation, which reveals unperturbed in-plane three-fold symmetry of the monolayer. Furthermore, a softening of $A_\mathrm{1g}$ out-of-plane lattice vibration is identified in the Raman spectroscopy, which is known to be insignificant for uniaxial strain. Concomitantly, PL mapping of our PVD monolayers demonstrates (i) larger strain occurs in the interior of the mono-domain islands compared to the edges and (ii) the absence of island-size dependence in the magnitude of induced strain. Our results demonstrate an effective path towards strain engineering of excitons by using growth substrates, which holds great promise as a building block for future optoelectronic applications.

We present results from the molecular dynamics (MD) simulation of surface-directed spinodal decomposition (SDSD) in binary fluid mixtures ($A+B$) with off-critical compositions. The aim is to elucidate the role of composition ratio in the early-time wetting kinetics under the influence of long-range surface potential. In our simulations, the attractive part of surface potential varies as $V(z)= -\epsilon_a/z^{n}$, with $\epsilon_{a}$ being the surface-potential strength. The surface prefers `$A$' species to form the wetting layer. Its thickness [$R_1(t)$] for the majority wetting (number of $A$-type particles [$N_A$] > number of $B$-type particles [$N_B$]), grows as a power-law with an exponent $1/(n+2)$. This is consistent with the early-time kinetics in the form of potential-dependent growth present in the Puri-Binder model. However, for minority wetting ($N_A$ < $N_B$), the growth exponent in $R_1(t)$ is less than $1/(n+2)$. Furthermore, on decreasing the field strength $\epsilon_{a}$, we recover $1/(n+2)$ for a minority wetting case. We provide phenomenological arguments to explain the early-time wetting kinetics for both cases., Comment: 11 pages, 7 figures

Recent experimental advances have made it possible to implement logical multi-qubit transversal gates on surface codes in a multitude of platforms. A transversal controlled-NOT (tCNOT) gate on two surface codes introduces correlated errors across the code blocks and thus requires modified decoding strategies compared to established methods of decoding surface code quantum memory (SCQM) or lattice surgery operations. In this work, we examine and benchmark the performance of three different decoding strategies for the tCNOT for scalable, fault-tolerant quantum computation. In particular, we present a low-complexity decoder based on minimum-weight perfect matching (MWPM) that achieves the same threshold as the SCQM MWPM decoder. We extend our analysis with a study of tailored decoding of a transversal teleportation circuit, along with a comparison between the performance of lattice surgery and transversal operations under Pauli and erasure noise models. Our investigation works towards systematic estimation of the cost of implementing large-scale quantum algorithms based on transversal gates in the surface code., Comment: 17 pages, 12 figures; v2 - minor clarifying changes, extended discussion in Sec. 3 and App. C1

Recently, a lot of effort has been devoted towards designing erasure qubits in which dominant physical noise excites leakage states whose population can be detected and returned to the qubit subspace. Interest in these erasure qubits has been driven by studies showing that the requirements for fault-tolerant quantum error correction are significantly relaxed when noise in every gate operation is dominated by erasures. However, these studies assume perfectly accurate erasure checks after every gate operation which generally come with undesirable time and hardware overhead costs. In this work, we investigate the consequences of using an imperfect but overhead-efficient erasure check for fault-tolerant quantum error correction with the surface code. We show that, under physically reasonable assumptions on the imperfect erasure checks, the threshold error rate is still at least over twice that for Pauli noise. We also study the impact of imperfect erasure checks on the effective error distance and find that it degrades the effective distance under a general error model in which a qubit suffers from depolarizing noise when interacting with a leaked qubit. We then identify a more restrictive but realistic noise model for a qubit that interacts with a leaked qubit, under which the effective error distance is twice that for Pauli noise. We apply our analysis to recently proposed superconducting dual-rail erasure qubits and show that achieving good performance surface code quantum memories with relaxed system requirements is possible., Comment: 19 pages, 15 figures

We analyse the impact of dark matter density spike around the Milky Way's supermassive black hole (SMBH), Sgr A$^*$, in probing the Bino-dominated neutralino dark matter (DM) $\tilde \chi_1^0$ within the MSSM, which typically produces relatively faint signals in the conventional DM halos. In particular, we explore the indirect search prospects of sub-TeV Bino-Higgsino and Bino-Wino-Higgsino DM in the MSSM, consistent with the supersymmetric predictions required to explain the anomalous magnetic moment of the muon. Typical over-abundance of Bino DM is ameliorated with slepton and/or Wino coannihilations. The lightest neutralino, thus, may be associated with a compressed supersymmetric particle spectrum, which, in general, is difficult to probe at conventional LHC searches. Similarly, for a rather tiny Higgsino mixing, $\tilde \chi_1^0$ does not offer much prospect to assess its predictions at dark matter direct detection searches. Accommodating the inclusive effects of density spike, here, we present the requisite boost factor to facilitate $\gamma-$ray searches of Bino-dominated DM in the MSSM, especially focusing on the Fermi-LAT and HESS observations., Comment: 29 pages, 11 figures, comments are welcome

This paper introduces long-context Granite code models that support effective context windows of up to 128K tokens. Our solution for scaling context length of Granite 3B/8B code models from 2K/4K to 128K consists of a light-weight continual pretraining by gradually increasing its RoPE base frequency with repository-level file packing and length-upsampled long-context data. Additionally, we also release instruction-tuned models with long-context support which are derived by further finetuning the long context base models on a mix of permissively licensed short and long-context instruction-response pairs. While comparing to the original short-context Granite code models, our long-context models achieve significant improvements on long-context tasks without any noticeable performance degradation on regular code completion benchmarks (e.g., HumanEval). We release all our long-context Granite code models under an Apache 2.0 license for both research and commercial use.

Data analytics is essential for extracting valuable insights from data that can assist organizations in making effective decisions. We introduce InsightBench, a benchmark dataset with three key features. First, it consists of 100 datasets representing diverse business use cases such as finance and incident management, each accompanied by a carefully curated set of insights planted in the datasets. Second, unlike existing benchmarks focusing on answering single queries, InsightBench evaluates agents based on their ability to perform end-to-end data analytics, including formulating questions, interpreting answers, and generating a summary of insights and actionable steps. Third, we conducted comprehensive quality assurance to ensure that each dataset in the benchmark had clear goals and included relevant and meaningful questions and analysis. Furthermore, we implement a two-way evaluation mechanism using LLaMA-3 as an effective, open-source evaluator to assess agents' ability to extract insights. We also propose AgentPoirot, our baseline data analysis agent capable of performing end-to-end data analytics. Our evaluation on InsightBench shows that AgentPoirot outperforms existing approaches (such as Pandas Agent) that focus on resolving single queries. We also compare the performance of open- and closed-source LLMs and various evaluation strategies. Overall, this benchmark serves as a testbed to motivate further development in comprehensive automated data analytics.

AI Infrastructure plays a key role in the speed and cost-competitiveness of developing and deploying advanced AI models. The current demand for powerful AI infrastructure for model training is driven by the emergence of generative AI and foundational models, where on occasion thousands of GPUs must cooperate on a single training job for the model to be trained in a reasonable time. Delivering efficient and high-performing AI training requires an end-to-end solution that combines hardware, software and holistic telemetry to cater for multiple types of AI workloads. In this report, we describe IBM's hybrid cloud infrastructure that powers our generative AI model development. This infrastructure includes (1) Vela: an AI-optimized supercomputing capability directly integrated into the IBM Cloud, delivering scalable, dynamic, multi-tenant and geographically distributed infrastructure for large-scale model training and other AI workflow steps and (2) Blue Vela: a large-scale, purpose-built, on-premises hosting environment that is optimized to support our largest and most ambitious AI model training tasks. Vela provides IBM with the dual benefit of high performance for internal use along with the flexibility to adapt to an evolving commercial landscape. Blue Vela provides us with the benefits of rapid development of our largest and most ambitious models, as well as future-proofing against the evolving model landscape in the industry. Taken together, they provide IBM with the ability to rapidly innovate in the development of both AI models and commercial offerings., Comment: Corresponding Authors: Talia Gershon, Seetharami Seelam,Brian Belgodere, Milton Bonilla

Nonlinear optical materials of atomic thickness--such as non-centrosymmetric 2H transition metal dichalcogenide monolayers--have a second order nonlinear susceptibility ($\chi^{(2)}$) whose intensity can be tuned by strain. However, whether $\chi^{(2)}$ is enhanced or reduced by tensile strain is a subject of conflicting reports. Here, we grow high-quality MoSe$_2$ monolayers under controlled biaxial strain created by two different substrates, and study their linear and non-linear optical responses with a combination of experimental and theoretical approaches. A 15-fold overall enhancement in second harmonic generation (SHG) intensity is observed on MoSe$_2$ monolayers grown on SiO$_2$ when compared to its value when on a Si$_3$N$_4$ substrate. A seven-fold enhancement was ascertained to substrate interference, and a factor of two to the enhancement of $\chi^{(2)}$ arising from biaxial strain: substrate interference and strain are independent handles to engineer the SHG strength of non-centrosymmetric 2D materials.

In the domain growth process, small structures gradually vanish, leaving behind larger ones. We investigate spectral energy transfers in two standard models for domain growth: (a) the {\it Cahn-Hilliard} (CH) equation with conserved dynamics, and (b) the {\it time-dependent Ginzburg-Landau} (TDGL) equation with non-conserved dynamics. The nonlinear terms in these equations dissipate fluctuations and facilitate energy transfers among Fourier modes. In the TDGL equation, only the $\phi(\mathbf{k} = 0, t)$ mode survives, and the order parameter $\phi(\mathbf{r},t)$ approaches a uniform state with $\phi = +1$ or $-1$. On the other hand, there is no dynamics of the $\phi(\mathbf{k} = 0, t)$ mode in the CH equation due to the conservation law, highlighting the different dynamics of these equations.

Quantum programs are typically developed using quantum Software Development Kits (SDKs). The rapid advancement of quantum computing necessitates new tools to streamline this development process, and one such tool could be Generative Artificial intelligence (GenAI). In this study, we introduce and use the Qiskit HumanEval dataset, a hand-curated collection of tasks designed to benchmark the ability of Large Language Models (LLMs) to produce quantum code using Qiskit - a quantum SDK. This dataset consists of more than 100 quantum computing tasks, each accompanied by a prompt, a canonical solution, a comprehensive test case, and a difficulty scale to evaluate the correctness of the generated solutions. We systematically assess the performance of a set of LLMs against the Qiskit HumanEval dataset's tasks and focus on the models ability in producing executable quantum code. Our findings not only demonstrate the feasibility of using LLMs for generating quantum code but also establish a new benchmark for ongoing advancements in the field and encourage further exploration and development of GenAI-driven tools for quantum code generation.

Quantum control of a linear oscillator using a static dispersive coupling to a nonlinear ancilla underpins a wide variety of experiments in circuit QED. Extending this control to more than one oscillator while minimizing the required connectivity to the ancilla would enable hardware-efficient multi-mode entanglement and measurements. We show that the spectrum of an ancilla statically coupled to a single mode can be made to depend on the joint photon number in two modes by applying a strong parametric beamsplitter coupling between them. This `joint-photon number-splitting' regime extends single-oscillator techniques to two-oscillator control, which we use to realize a hardware-efficient erasure check for a dual-rail qubit encoded in two superconducting cavities. By leveraging the beamsplitter coupling already required for single-qubit gates, this scheme permits minimal connectivity between circuit elements. Furthermore, the flexibility to choose the pulse shape allows us to limit the susceptibility to different error channels. We use this scheme to detect leakage errors with a missed erasure fraction of $(9.0 \pm 0.5)\times10^{-4}$, while incurring an erasure rate of $2.92 \pm 0.01\%$ and a Pauli error rate of $0.31 \pm 0.01\%$, both of which are dominated by cavity errors., Comment: Updated references

We study the phase separation kinetics of miktoarm star polymer (MSP) melts and blends with diverse architectures using dissipative particle dynamics simulations. Our study focuses on symmetric and asymmetric miktoarm star polymer (SMSP/AMSP) mixtures based on arm composition and number. For a fixed MSP chain size, the characteristic microphase-separated domains initially show diffusive growth with a growth exponent $\phi \sim 1/3$ for both melts that gradually crossover to saturation at late times. The simulation results demonstrate that the evolution morphology of SMSP melts exhibits perfect dynamic scaling with varying arm numbers; the time scale follows a power-law decay with an exponent $\theta \simeq 1$ as the number of arms increases. The structural constraints on AMSP melts cause the domain growth rate to decrease as the number of one type of arms increases while their length remains fixed. This increase in the number of arms for AMSP corresponds to increased off-criticality. The saturation length in AMSP follows a power law increase with an exponent $\lambda \simeq 2/3$ as off-criticality decreases. Additionally, macrophase separation kinetics in SMSP/AMSP blends show a transition from viscous ($\phi \sim 1$) to inertial ($\phi \sim 2/3$) hydrodynamic growth regimes at late times; this exhibits the same dynamical universality class as linear polymer blends, with slight deviations at early stages.

Electronic design engineers are challenged to find relevant information efficiently for a myriad of tasks within design construction, verification and technology development. Large language models (LLM) have the potential to help improve productivity by serving as conversational agents that effectively function as subject-matter experts. In this paper we demonstrate Ask-EDA, a chat agent designed to serve as a 24x7 expert available to provide guidance to design engineers. Ask-EDA leverages LLM, hybrid retrieval augmented generation (RAG) and abbreviation de-hallucination (ADH) techniques to deliver more relevant and accurate responses. We curated three evaluation datasets, namely q2a-100, cmds-100 and abbr-100. Each dataset is tailored to assess a distinct aspect: general design question answering, design command handling and abbreviation resolution. We demonstrated that hybrid RAG offers over a 40% improvement in Recall on the q2a-100 dataset and over a 60% improvement on the cmds-100 dataset compared to not using RAG, while ADH yields over a 70% enhancement in Recall on the abbr-100 dataset. The evaluation results show that Ask-EDA can effectively respond to design-related inquiries., Comment: Accepted paper at The First IEEE International Workshop on LLM-Aided Design, 2024 (LAD 24)

Code Large Language Models (Code LLMs) have emerged as powerful tools, revolutionizing the software development landscape by automating the coding process and reducing time and effort required to build applications. This paper focuses on training Code LLMs to specialize in the field of quantum computing. We begin by discussing the unique needs of quantum computing programming, which differ significantly from classical programming approaches or languages. A Code LLM specializing in quantum computing requires a foundational understanding of quantum computing and quantum information theory. However, the scarcity of available quantum code examples and the rapidly evolving field, which necessitates continuous dataset updates, present significant challenges. Moreover, we discuss our work on training Code LLMs to produce high-quality quantum code using the Qiskit library. This work includes an examination of the various aspects of the LLMs used for training and the specific training conditions, as well as the results obtained with our current models. To evaluate our models, we have developed a custom benchmark, similar to HumanEval, which includes a set of tests specifically designed for the field of quantum computing programming using Qiskit. Our findings indicate that our model outperforms existing state-of-the-art models in quantum computing tasks. We also provide examples of code suggestions, comparing our model to other relevant code LLMs. Finally, we introduce a discussion on the potential benefits of Code LLMs for quantum computing computational scientists, researchers, and practitioners. We also explore various features and future work that could be relevant in this context.

Large language models (LLMs) are already being piloted for clinical use in hospital systems like NYU Langone, Dana-Farber and the NHS. A proposed deployment use case is psychotherapy, where a LLM-powered chatbot can treat a patient undergoing a mental health crisis. Deployment of LLMs for mental health response could hypothetically broaden access to psychotherapy and provide new possibilities for personalizing care. However, recent high-profile failures, like damaging dieting advice offered by the Tessa chatbot to patients with eating disorders, have led to doubt about their reliability in high-stakes and safety-critical settings. In this work, we develop an evaluation framework for determining whether LLM response is a viable and ethical path forward for the automation of mental health treatment. Our framework measures equity in empathy and adherence of LLM responses to motivational interviewing theory. Using human evaluation with trained clinicians and automatic quality-of-care metrics grounded in psychology research, we compare the responses provided by peer-to-peer responders to those provided by a state-of-the-art LLM. We show that LLMs like GPT-4 use implicit and explicit cues to infer patient demographics like race. We then show that there are statistically significant discrepancies between patient subgroups: Responses to Black posters consistently have lower empathy than for any other demographic group (2%-13% lower than the control group). Promisingly, we do find that the manner in which responses are generated significantly impacts the quality of the response. We conclude by proposing safety guidelines for the potential deployment of LLMs for mental health response., Comment: EMNLP 2024 Findings

Reduced order modeling lowers the computational cost of solving PDEs by learning a low-order spatial representation from data and dynamically evolving these representations using manifold projections of the governing equations. While commonly used, linear subspace reduced-order models (ROMs) are often suboptimal for problems with a slow decay of Kolmogorov $n$-width, such as advection-dominated fluid flows at high Reynolds numbers. There has been a growing interest in nonlinear ROMs that use state-of-the-art representation learning techniques to accurately capture such phenomena with fewer degrees of freedom. We propose smooth neural field ROM (SNF-ROM), a nonlinear reduced modeling framework that combines grid-free reduced representations with Galerkin projection. The SNF-ROM architecture constrains the learned ROM trajectories to a smoothly varying path, which proves beneficial in the dynamics evaluation when the reduced manifold is traversed in accordance with the governing PDEs. Furthermore, we devise robust regularization schemes to ensure the learned neural fields are smooth and differentiable. This allows us to compute physics-based dynamics of the reduced system nonintrusively with automatic differentiation and evolve the reduced system with classical time-integrators. SNF-ROM leads to fast offline training as well as enhanced accuracy and stability during the online dynamics evaluation. Numerical experiments reveal that SNF-ROM is able to accelerate the full-order computation by up to $199\times$. We demonstrate the efficacy of SNF-ROM on a range of advection-dominated linear and nonlinear PDE problems where we consistently outperform state-of-the-art ROMs.

Large Language Models (LLMs) trained on code are revolutionizing the software development process. Increasingly, code LLMs are being integrated into software development environments to improve the productivity of human programmers, and LLM-based agents are beginning to show promise for handling complex tasks autonomously. Realizing the full potential of code LLMs requires a wide range of capabilities, including code generation, fixing bugs, explaining and documenting code, maintaining repositories, and more. In this work, we introduce the Granite series of decoder-only code models for code generative tasks, trained with code written in 116 programming languages. The Granite Code models family consists of models ranging in size from 3 to 34 billion parameters, suitable for applications ranging from complex application modernization tasks to on-device memory-constrained use cases. Evaluation on a comprehensive set of tasks demonstrates that Granite Code models consistently reaches state-of-the-art performance among available open-source code LLMs. The Granite Code model family was optimized for enterprise software development workflows and performs well across a range of coding tasks (e.g. code generation, fixing and explanation), making it a versatile all around code model. We release all our Granite Code models under an Apache 2.0 license for both research and commercial use., Comment: Corresponding Authors: Rameswar Panda, Ruchir Puri; Equal Contributors: Mayank Mishra, Matt Stallone, Gaoyuan Zhang

We use molecular dynamics simulations to study coarsening dynamics in achiral banana-shaped bent-core liquid crystals following a quench from the high concentration polar smectic (SmX) phase to lower concentrations that favor the exotic twist-bend (TB) phase. Our novel result is the identification of an intermediate splay-bend state emerging prior to the eventual TB phase. The latter coarsens via the annihilation of {\it beta lines} which are analogous to string defects in nematic liquid crystals. Our findings are relevant for a large class of chiral systems assembled from achiral entities.

The Kerr-cat qubit is a bosonic qubit in which multi-photon Schrodinger cat states are stabilized by applying a two-photon drive to an oscillator with a Kerr nonlinearity. The suppressed bit-flip rate with increasing cat size makes this qubit a promising candidate to implement quantum error correction codes tailored for noise-biased qubits. However, achieving strong light-matter interactions necessary for stabilizing and controlling this qubit has traditionally required strong microwave drives that heat the qubit and degrade its performance. In contrast, increasing the coupling to the drive port removes the need for strong drives at the expense of large Purcell decay. By integrating an effective band-block filter on-chip, we overcome this trade-off and realize a Kerr-cat qubit in a scalable 2D superconducting circuit with high coherence. This filter provides 30 dB of isolation at the qubit frequency with negligible attenuation at the frequencies required for stabilization and readout. We experimentally demonstrate quantum non-demolition readout fidelity of 99.6% for a cat with 8 photons. Also, to have high-fidelity universal control over this qubit, we combine fast Rabi oscillations with a new demonstration of the X(90) gate through phase modulation of the stabilization drive. Finally, the lifetime in this architecture is examined as a function of the cat size of up to 10 photons in the oscillator achieving a bit-flip time higher than 1 ms and only a linear decrease in the phase-flip time, in good agreement with the theoretical analysis of the circuit. Our qubit shows promise as a building block for fault-tolerant quantum processors with a small footprint.