Search

We survey undergraduate students at a large public university to understand the pecuniary and non-pecuniary factors driving their college major and career decisions with a focus on K-12 teaching. While the average student reports there is a 6% chance they will pursue teaching, almost 27% report a nonzero chance of working as a teacher in the future. Students, relative to existing statistics, generally believe they would earn substantially more in a non-teaching job (relative to a teaching job). We run a randomized information experiment where we provide students with information on the pecuniary and non-pecuniary job characteristics of teachers and non-teachers. This low-cost informational intervention impacts students' beliefs about their job characteristics if they were to work as a teacher or non-teacher, and increases the reported likelihood they will major or minor in education by 35% and pursue a job as a teacher or in education by 14%. Linking the survey data with administrative transcript records, we find that the intervention had small (and weak) impacts on the decision to minor in education in the subsequent year. Overall, our results indicate that students hold biased beliefs about their career prospects, they update these beliefs when provided with information, and that this information has limited impacts on their choices regarding studying and having a career in teaching.

This paper presents the concept of a tethered variable inertial attitude control mechanism for a modular jumping-limbed robot designed for planetary exploration in low-gravity environments. The system, named SPLITTER, comprises two sub-10 kg quadrupedal robots connected by a tether, capable of executing successive jumping gaits and stabilizing in-flight using inertial morphing technology. Through model predictive control (MPC), attitude control was demonstrated by adjusting the limbs and tether length to modulate the system's principal moments of inertia. Our results indicate that this control strategy allows the robot to stabilize during flight phases without needing traditional flywheel-based systems or relying on aerodynamics, making the approach mass-efficient and ideal for small-scale planetary robots' successive jumps. The paper outlines the dynamics, MPC formulation for inertial morphing, actuator requirements, and simulation results, illustrating the potential of agile exploration for small-scale rovers in low-gravity environments like the Moon or asteroids., Comment: Proceeding to IEEE Aerospace Conference 2025

Many documents, that we call templatized documents, are programmatically generated by populating fields in a visual template. Effective data extraction from these documents is crucial to supporting downstream analytical tasks. Current data extraction tools often struggle with complex document layouts, incur high latency and/or cost on large datasets, and often require significant human effort, when extracting tables or values given user-specified fields from documents. The key insight of our tool, TWIX, is to predict the underlying template used to create such documents, modeling the visual and structural commonalities across documents. Data extraction based on this predicted template provides a more principled, accurate, and efficient solution at a low cost. Comprehensive evaluations on 34 diverse real-world datasets show that uncovering the template is crucial for data extraction from templatized documents. TWIX achieves over 90% precision and recall on average, outperforming tools from industry: Textract and Azure Document Intelligence, and vision-based LLMs like GPT-4-Vision, by over 25% in precision and recall. TWIX scales easily to large datasets and is 734X faster and 5836X cheaper than vision-based LLMs for extracting data from a large document collection with 817 pages.

Developing reliable healthcare AI models requires training with representative and diverse data. In imbalanced datasets, model performance tends to plateau on the more prevalent classes while remaining low on less common cases. To overcome this limitation, we propose DiffUltra, the first generative AI technique capable of synthesizing realistic Lung Ultrasound (LUS) images with extensive lesion variability. Specifically, we condition the generative AI by the introduced Lesion-anatomy Bank, which captures the lesion's structural and positional properties from real patient data to guide the image synthesis.We demonstrate that DiffUltra improves consolidation detection by 5.6% in AP compared to the models trained solely on real patient data. More importantly, DiffUltra increases data diversity and prevalence of rare cases, leading to a 25% AP improvement in detecting rare instances such as large lung consolidations, which make up only 10% of the dataset., Comment: Accepted by ISBI 2025

This work introduces a method for correcting the output distribution of a quantum computer that does not require encoding of the logical qubits into more physical qubits. Thus, it avoids the encoding overhead of standard quantum error correction codes. If the noise affecting the circuit is a Pauli channel (we can bias the noise with twirling), the ideal output distribution and noisy distribution in the standard basis are related by a stochastic matrix. We prove that this matrix has a circulant block structure. Thus, the ideal distribution can be retrieved from the noisy output distribution by applying the Fast Fourier Transform. Moreover, due to its circulant structure, characterization of this matrix can be achieved by sampling a single circuit. The results are corroborated with quantum hardware executions consisting of 20-qubit and 30-qubit GHZ state preparation, 5-qubit Grover, 6-qubit and 10-qubit quantum phase estimation, and 10-qubit and 20-qubit Dicke state preparation circuits. The correction process dramatically improves the accuracies of the output distributions. For 30-qubit GHZ state preparation, a corrected distribution fidelity of 97.7% is achieved from an initial raw fidelity of 23.2%., Comment: Comments are welcome!

Matched-filtering searches for gravitational-wave signals from compact binary mergers employ template banks which are a collection of modeled waveforms described by unique intrinsic parameters. We present two banks designed for low-latency and archive sub-solar mass (SSM) searches in data from the fourth observing run of LIGO-Virgo-KAGRA, and demonstrate the efficacy of the banks via simulated signals. Further, we introduce a set of modifications to the geometric, manifold algorithm that allow the method to work for exceedingly low component masses necessary for SSM bank production. The archive search bank contains a total of $3,452,006$ templates, and covers a mass parameter space of $0.2$ to $10\ M_\odot$ in the larger component and $0.2$ to $1.0\ M_\odot$ in the smaller component, the spin parameter space of $-0.9$ to $0.9$ for masses above $0.5$ $M_\odot$ and $-0.1$ to $0.1$ for masses below $0.5$ $M_\odot$, and the mass ratio parameter space of $1$ to $10$. The PSD used was from a week of the first half of the fourth observing run of Advanced LIGO, Advanced Virgo, and KAGRA, and the low frequency cutoff was set to $45$ Hz with a maximum waveform duration of $128$ seconds. The bank simulations performed using SBank have shown that the banks presented in this paper have sufficient efficacy for use in their respective searches.

Recent advances in multiplex imaging have enabled researchers to locate different types of cells within a tissue sample. This is especially relevant for tumor immunology, as clinical regimes corresponding to different stages of disease or responses to treatment may manifest as different spatial arrangements of tumor and immune cells. Spatial point pattern modeling can be used to partition multiplex tissue images according to these regimes. To this end, we propose a two-stage approach: first, local intensities and pair correlation functions are estimated from the spatial point pattern of cells within each image, and the pair correlation functions are reduced in dimension via spectral decomposition of the covariance function. Second, the estimates are clustered in a Bayesian hierarchical model with spatially-dependent cluster labels. The clusters correspond to regimes of interest that are present across subjects; the cluster labels segment the spatial point patterns according to those regimes. Through Markov Chain Monte Carlo sampling, we jointly estimate and quantify uncertainty in the cluster assignment and spatial characteristics of each cluster. Simulations demonstrate the performance of the method, and it is applied to a set of multiplex immunofluorescence images of diseased pancreatic tissue., Comment: 13 pages, 5 figures, 1 table

Here, we present the outcomes from the second Large Language Model (LLM) Hackathon for Applications in Materials Science and Chemistry, which engaged participants across global hybrid locations, resulting in 34 team submissions. The submissions spanned seven key application areas and demonstrated the diverse utility of LLMs for applications in (1) molecular and material property prediction; (2) molecular and material design; (3) automation and novel interfaces; (4) scientific communication and education; (5) research data management and automation; (6) hypothesis generation and evaluation; and (7) knowledge extraction and reasoning from scientific literature. Each team submission is presented in a summary table with links to the code and as brief papers in the appendix. Beyond team results, we discuss the hackathon event and its hybrid format, which included physical hubs in Toronto, Montreal, San Francisco, Berlin, Lausanne, and Tokyo, alongside a global online hub to enable local and virtual collaboration. Overall, the event highlighted significant improvements in LLM capabilities since the previous year's hackathon, suggesting continued expansion of LLMs for applications in materials science and chemistry research. These outcomes demonstrate the dual utility of LLMs as both multipurpose models for diverse machine learning tasks and platforms for rapid prototyping custom applications in scientific research., Comment: Updating author information, the submission remains largely unchanged. 98 pages total

Gravitational lensing by compact, small-scale intervening masses causes frequency-dependent distortions to gravitational-wave events. The optimal signal-to-noise ratio (SNR) is often used as a proxy for the detectability of exotic signals in gravitational-wave searches. In reality, the detectability of such signals in a matched-filtering search requires comprehensive consideration of match-filtered SNR, signal-consistency test value, and other factors. In this work, we investigate for the first time the detectability of lensed gravitational waves from compact binary coalescences with a match-filtering search pipeline, GstLAL. Contrary to expectations from the optimal-SNR approximation approach, we show that the strength of a signal (i.e., higher optimal SNR) does not necessarily result in higher detectability. We also demonstrate that lensed gravitational waves with wave optics effects can suffer significantly, from $~90\%$ (unlensed) to $<1\%$ (lensed) detection efficiency, due to downranking by the signal-consistency test values. These findings stress the need to extend current template banks to effectively search for lensed gravitational waves and to reassess current constraints on compact dark matter scenarios., Comment: 15 pages, 7 figures

Understanding spatial relations is a crucial cognitive ability for both humans and AI. While current research has predominantly focused on the benchmarking of text-to-image (T2I) models, we propose a more comprehensive evaluation that includes \textit{both} T2I and Large Language Models (LLMs). As spatial relations are naturally understood in a visuo-spatial manner, we develop an approach to convert LLM outputs into an image, thereby allowing us to evaluate both T2I models and LLMs \textit{visually}. We examined the spatial relation understanding of 8 prominent generative models (3 T2I models and 5 LLMs) on a set of 10 common prepositions, as well as assess the feasibility of automatic evaluation methods. Surprisingly, we found that T2I models only achieve subpar performance despite their impressive general image-generation abilities. Even more surprisingly, our results show that LLMs are significantly more accurate than T2I models in generating spatial relations, despite being primarily trained on textual data. We examined reasons for model failures and highlight gaps that can be filled to enable more spatially faithful generations.

Streaming systems are present throughout modern applications, processing continuous data in real-time. Existing streaming languages have a variety of semantic models and guarantees that are often incompatible. Yet all these languages are considered "streaming" -- what do they have in common? In this paper, we identify two general yet precise semantic properties: streaming progress and eager execution. Together, they ensure that streaming outputs are deterministic and kept fresh with respect to streaming inputs. We formally define these properties in the context of Flo, a parameterized streaming language that abstracts over dataflow operators and the underlying structure of streams. It leverages a lightweight type system to distinguish bounded streams, which allow operators to block on termination, from unbounded ones. Furthermore, Flo provides constructs for dataflow composition and nested graphs with cycles. To demonstrate the generality of our properties, we show how key ideas from representative streaming and incremental computation systems -- Flink, LVars, and DBSP -- have semantics that can be modeled in Flo and guarantees that map to our properties.

Recent works have shown a surprising result: a small fraction of Large Language Model (LLM) parameter outliers are disproportionately important to the quality of the model. LLMs contain billions of parameters, so these small fractions, such as 0.01%, translate to hundreds of thousands of parameters. In this work, we present an even more surprising finding: Pruning as few as a single parameter can destroy an LLM's ability to generate text -- increasing perplexity by 3 orders of magnitude and reducing zero-shot accuracy to guessing. We propose a data-free method for identifying such parameters, termed super weights, using a single forward pass through the model. We additionally find that these super weights induce correspondingly rare and large activation outliers, termed super activations. When preserved with high precision, super activations can improve simple round-to-nearest quantization to become competitive with state-of-the-art methods. For weight quantization, we similarly find that by preserving the super weight and clipping other weight outliers, round-to-nearest quantization can scale to much larger block sizes than previously considered. To facilitate further research into super weights, we provide an index of super weight coordinates for common, openly available LLMs.

High-performance analog and mixed-signal (AMS) circuits are mainly full-custom designed, which is time-consuming and labor-intensive. A significant portion of the effort is experience-driven, which makes the automation of AMS circuit design a formidable challenge. Large language models (LLMs) have emerged as powerful tools for Electronic Design Automation (EDA) applications, fostering advancements in the automatic design process for large-scale AMS circuits. However, the absence of high-quality datasets has led to issues such as model hallucination, which undermines the robustness of automatically generated circuit designs. To address this issue, this paper introduces AMSnet-KG, a dataset encompassing various AMS circuit schematics and netlists. We construct a knowledge graph with annotations on detailed functional and performance characteristics. Facilitated by AMSnet-KG, we propose an automated AMS circuit generation framework that utilizes the comprehensive knowledge embedded in LLMs. We first formulate a design strategy (e.g., circuit architecture using a number of circuit components) based on required specifications. Next, matched circuit components are retrieved and assembled into a complete topology, and transistor sizing is obtained through Bayesian optimization. Simulation results of the netlist are fed back to the LLM for further topology refinement, ensuring the circuit design specifications are met. We perform case studies of operational amplifier and comparator design to verify the automatic design flow from specifications to netlists with minimal human effort. The dataset used in this paper will be open-sourced upon publishing of this paper.

Biomedical research requires large, diverse samples to produce unbiased results. Automated methods for matching variables across datasets can accelerate this process. Research in this area has been limited, primarily focusing on lexical matching and ontology based semantic matching. We aimed to develop new methods, leveraging large language models (LLM) and ensemble learning, to automate variable matching. Methods: We utilized data from two GERAS cohort (European and Japan) studies to develop variable matching methods. We first manually created a dataset by matching 352 EU variables with 1322 candidate JP variables, where matched variable pairs were positive and unmatched pairs were negative instances. Using this dataset, we developed and evaluated two types of natural language processing (NLP) methods, which matched variables based on variable labels and definitions from data dictionaries: (1) LLM-based and (2) fuzzy matching. We then developed an ensemble-learning method, using the Random Forest model, to integrate individual NLP methods. RF was trained and evaluated on 50 trials. Each trial had a random split (4:1) of training and test sets, with the model's hyperparameters optimized through cross-validation on the training set. For each EU variable, 1322 candidate JP variables were ranked based on NLP-derived similarity scores or RF's probability scores, denoting their likelihood to match the EU variable. Ranking performance was measured by top-n hit ratio (HRn) and mean reciprocal rank (MRR). Results:E5 performed best among individual methods, achieving 0.90 HR-30 and 0.70 MRR. RF performed better than E5 on all metrics over 50 trials (P less than 0.001) and achieved an average HR 30 of 0.98 and MRR of 0.73. LLM-derived features contributed most to RF's performance. One major cause of errors in automatic variable matching was ambiguous variable definitions within data dictionaries., Comment: 32 pages, 2 figures

Quasi-periodic eruptions (QPEs) are recurring bursts of soft X-rays from the nuclei of galaxies. Their physical origin is currently a subject of debate, with models typically invoking an orbiter around a massive black hole or disk instabilities. Here we present and analyze the temporal and spectral evolution of the QPE source eRO-QPE2 over 3.5 years. We find that eRO-QPE2 1) is remarkably stable over the entire 3.5-year temporal baseline in its eruption peak luminosity, eruption temperature, quiescent temperature, and quiescent luminosity, 2) has a stable mean eruption recurrence time of 2.35 hours, with marginal ($\sim$2$\sigma$) evidence for a $0.1$ hour reduction over the 3.5 yr period, and 3) has a long-short variation in its recurrence time in August 2020, but this pattern is absent from all subsequent observations. The stability of its peak eruption luminosity and that of the quiescent state are notably dissimilar from three previously tracked QPEs (GSN069, eRO-QPE1, eRO-QPE3), which show declines in eruption and quiescent flux over comparable temporal baselines. This stability is even more pronounced in eRO-QPE2 due to its 2.4 hour average recurrence time compared to GSN-069's 9 hour, eRO-QPE1's 16 hour, and eRO-QPE3's 20 hour recurrence times, i.e., this system has undergone 4-8 times more cycles than these other systems over the 3.5 years of observations. We discuss the implications of these observations within the context of some proposed extreme mass ratio inspiral (EMRI) models., Comment: Under review (ApJ)

Concatenated bit-interleaved and multilevel coded modulation with outer Reed--Solomon codes, inner Chase-algorithm-based soft-decision-decoded Bose--Ray-Chaudhuri--Hocquenghem codes, and four-level pulse amplitude modulation is considered. A semi-analytical formula is derived for estimating the decoded frame error rate (FER) at the output the additive white Gaussian noise channel, obviating the need for time-consuming Monte Carlo simulations. The formula is used to search a large space of codes (including the KP4 code) to find those achieving good trade-offs among performance (measured by the gap to the constrained Shannon limit at $10^{-13}$ FER), complexity (measured by the number of elementary decoder operations), and latency (measured by overall block length)., Comment: Submitted to IEEE Journal of Lightwave Technology (JLT)

In limbed robotics, end-effectors must serve dual functions, such as both feet for locomotion and grippers for grasping, which presents design challenges. This paper introduces a multi-modal end-effector capable of transitioning between flat and line foot configurations while providing grasping capabilities. MAGPIE integrates 8-axis force sensing using proposed mechanisms with hall effect sensors, enabling both contact and tactile force measurements. We present a computational design framework for our sensing mechanism that accounts for noise and interference, allowing for desired sensitivity and force ranges and generating ideal inverse models. The hardware implementation of MAGPIE is validated through experiments, demonstrating its capability as a foot and verifying the performance of the sensing mechanisms, ideal models, and gated network-based models.

This paper presents a novel approach through the design and implementation of Cycloidal Quasi-Direct Drive actuators for legged robotics. The cycloidal gear mechanism, with its inherent high torque density and mechanical robustness, offers significant advantages over conventional designs. By integrating cycloidal gears into the Quasi-Direct Drive framework, we aim to enhance the performance of legged robots, particularly in tasks demanding high torque and dynamic loads, while still keeping them lightweight. Additionally, we develop a torque estimation framework for the actuator using an Actuator Network, which effectively reduces the sim-to-real gap introduced by the cycloidal drive's complex dynamics. This integration is crucial for capturing the complex dynamics of a cycloidal drive, which contributes to improved learning efficiency, agility, and adaptability for reinforcement learning.

In this work we compare the generative behavior at the next token prediction level in several language models by comparing them to human productions in the cloze task. We find that while large models trained for longer are typically better estimators of human productions, but they reliably under-estimate the probabilities of human responses, over-rank rare responses, under-rank top responses, and produce highly distinct semantic spaces. Altogether, this work demonstrates in a tractable, interpretable domain that LM generations can not be used as replacements of or models of the cloze task.

A binary extreme-mass-ratio inspiral (b-EMRI) is a hierarchical triple system consisting of a stellar-mass binary black hole (BBH) orbiting a central Kerr supermassive black hole (SMBH). Although predicted by several astrophysical models, b-EMRIs pose a challenge in waveform modeling due to their complex three-body dynamics and strong relativistic effects. Here we take advantage of the hierarchical nature of b-EMRI systems to transform the internal motion of the small binary into global trajectories around the SMBH. This allows us to use black hole perturbation theory to calculate both the low-frequency gravitational waveform due to its EMRI nature and the high-frequency waveform generated by the inner motion of the BBH. When the inner binary's separation vanishes, our calculation recovers the standard relativistic adiabatic EMRI waveform. Furthermore, by including the high-frequency perturbation, we find a correction to the waveform as large as the adiabatic order when the frequency matches the quasinormal modes (QNMs) of the SMBH, therefore supporting an earlier proof-of-concept study claiming that the small BBH can resonantly excite the QNMs of the SMBH. More importantly, we find that b-EMRIs can evolve faster than regular EMRIs due to this resonant dissipation through the high-frequency modes. These characteristics distinguish b-EMRI waveform templates from regular EMRI templates for future space-based gravitational-wave detectors., Comment: 26 pages, 7 figures, submitted to PRD

The opening of the gravitational wave window has significantly enhanced our capacity to explore the universe's most extreme and dynamic sector. In the mHz frequency range, a diverse range of compact objects, from the most massive black holes at the farthest reaches of the Universe to the lightest white dwarfs in our cosmic backyard, generate a complex and dynamic symphony of gravitational wave signals. Once recorded by gravitational wave detectors, these unique fingerprints have the potential to decipher the birth and growth of cosmic structures over a wide range of scales, from stellar binaries and stellar clusters to galaxies and large-scale structures. The TianQin space-borne gravitational wave mission is scheduled for launch in the 2030s, with an operational lifespan of five years. It will facilitate pivotal insights into the history of our universe. This document presents a concise overview of the detectable sources of TianQin, outlining their characteristics, the challenges they present, and the expected impact of the TianQin observatory on our understanding of them., Comment: TianQin Gravitational Wave Whitepaper, 72 pages, 30 figures

We show that quantum advantage in distributed sensing can be achieved with noisy quantum networks. When using depolarized GHZ states as the probe, we derive a closed-form fidelity threshold to achieve advantage over the optimal local sensing strategy. The threshold indicates that while entanglement is needed for this quantum advantage, genuine multipartite entanglement is generally unnecessary. We further explore the impacts from imperfect local entanglement generation and local measurement constraint, and our results imply that the quantum advantage is more robust against quantum network imperfections than local operation errors. Finally, we demonstrate that the quantum advantage in distributed sensing can be achieved with a three-node quantum network using practical protocol stacks through simulations with SeQUeNCe, an open-source, customizable quantum network simulator., Comment: Updated references, added copyright statement, modified format

Wide, highly eccentric ($e>0.9$) compact binaries can naturally arise as progenitors of gravitational wave (GW) mergers. These systems are expected to have a significant population in the mHz band (e.g., $\sim 3-45$ detectable stellar-mass binary black holes with $e>0.9$ in the Milky Way), with their GW signals characterized by "repeated bursts" emitted upon each pericenter passage. In this study, we show that the detection of mHz GW signals from highly eccentric stellar mass binaries in the local universe can strongly constrain their orbital parameters. Specifically, it can achieve a relative measurement error of $\sim 10^{-6}$ for orbital frequency and $\sim 1\%$ for eccentricity (as $1-e$) in most of the detectable cases. On the other hand, the binary's mass ratio, distance, and intrinsic orbital orientation may be less precisely determined due to degeneracies in the GW waveform. We also perform mock LISA data analysis to evaluate the realistic detectability of highly eccentric compact binaries. Our results show that highly eccentric systems could be efficiently identified when multiple GW sources and stationary Gaussian instrumental noise are present in the detector output. This work highlights the potential of extracting the signal of "bursting'' LISA sources to provide valuable insights into their orbital evolution, surrounding environment, and formation channels., Comment: 17+9 pages, 6+3 figures. Accepted by PRD

It has been shown in Phys. Rev. Lett., 108 170402 (2012) (https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.108.170402), that quantum tunneling is instantaneous using a time-of-arrival (TOA) operator constructed by Weyl quantization of the classical TOA. However, there are infinitely many possible quantum images of the classical TOA, leaving it unclear if one is uniquely preferred over the others. This raises the question on whether instantaneous tunneling time is simply an artifact of the chosen ordering rule. Here, we demonstrate that tunneling time vanishes for all possible quantum images of the classical arrival time, irrespective of the ordering rule between the position and momentum observables. The result still holds for TOA-operators that are constructed independent of canonical quantization, while still imposing the correct algebra defined by the time-energy canonical commutation relation.

On the 29th of May 2023, the LIGO-Virgo-KAGRA Collaboration observed a compact binary coalescence event consistent with a neutron star-black hole merger, though the heavier object of mass 2.5-4.5 $M_\odot$ would fall into the purported lower mass gap. An alternative explanation for apparent observations of events in this mass range has been suggested as strongly gravitationally lensed binary neutron stars. In this scenario, magnification would lead to the source appearing closer and heavier than it really is. Here, we investigate the chances and possible consequences for the GW230529 event to be gravitationally lensed. We find this would require high magnifications and we obtain low rates for observing such an event, with a relative fraction of lensed versus unlensed observed events of $2 \times 10^{-3}$ at most. When comparing the lensed and unlensed hypotheses accounting for the latest rates and population model, we find a 1/58 chance of lensing, disfavoring this option. Moreover, when the magnification is assumed to be strong enough to bring the mass of the heavier binary component below the standard limits on neutron star masses, we find high probability for the lighter object to have a sub-solar mass, making the binary even more exotic than a mass-gap neutron star-black hole system. Even when the secondary is not sub-solar, its tidal deformability would likely be measurable, which is not the case for GW230529. Finally, we do not find evidence for extra lensing signatures such as the arrival of additional lensed images, type-II image dephasing, or microlensing. Therefore, we conclude it is unlikely for GW230529 to be a strongly gravitationally lensed binary neutron star signal., Comment: 15 pages, 11 figures

Heralded entanglement generation between nodes of a future quantum internet is a fundamental operation that unlocks the potential for quantum communication. In this paper, we propose a decentralized synchronization protocol that operates at the classical control-plane of the link layer, to navigate the coordination challenges of generating heralded entanglement across few-qubit quantum network nodes. Additionally, with quantum network simulations using NetSquid, we show that our protocol achieves lower entanglement request latencies than a naive distributed queue approach. We observe a sixfold reduction in average request latency growth as the number of quantum network links increases. The Eventual Synchronization Protocol (ESP) allows nodes to coordinate on heralded entanglement generation in a scalable manner within multi-peer quantum networks. To the best of our knowledge, this is the first decentralized synchronization protocol for managing heralded entanglement requests., Comment: accepted by the 20th International Conference on Network and Service Management (CNSM), 2024

Multimodal fusion promises better pancreas segmentation. However, where to perform fusion in models is still an open question. It is unclear if there is a best location to fuse information when analyzing pairs of imperfectly aligned images. Two main alignment challenges in this pancreas segmentation study are 1) the pancreas is deformable and 2) breathing deforms the abdomen. Even after image registration, relevant deformations are often not corrected. We examine how early through late fusion impacts pancreas segmentation. We used 353 pairs of T2-weighted (T2w) and T1-weighted (T1w) abdominal MR images from 163 subjects with accompanying pancreas labels. We used image registration (deeds) to align the image pairs. We trained a collection of basic UNets with different fusion points, spanning from early to late, to assess how early through late fusion influenced segmentation performance on imperfectly aligned images. We assessed generalization of fusion points on nnUNet. The single-modality T2w baseline using a basic UNet model had a Dice score of 0.73, while the same baseline on the nnUNet model achieved 0.80. For the basic UNet, the best fusion approach occurred in the middle of the encoder (early/mid fusion), which led to a statistically significant improvement of 0.0125 on Dice score compared to the baseline. For the nnUNet, the best fusion approach was na\"ive image concatenation before the model (early fusion), which resulted in a statistically significant Dice score increase of 0.0021 compared to baseline. Fusion in specific blocks can improve performance, but the best blocks for fusion are model specific, and the gains are small. In imperfectly registered datasets, fusion is a nuanced problem, with the art of design remaining vital for uncovering potential insights. Future innovation is needed to better address fusion in cases of imperfect alignment of abdominal image pairs., Comment: 13.5 pages of manuscript content

Hazardous asteroid has been one of the concerns for humankind as fallen asteroid on earth could cost a huge impact on the society.Monitoring these objects could help predict future impact events, but such efforts are hindered by the large numbers of objects that pass in the Earth's vicinity. The aim of this project is to use machine learning and deep learning to accurately classify hazardous asteroids. A total of ten methods which consist of five machine learning algorithms and five deep learning models are trained and evaluated to find the suitable model that solves the issue. We experiment on two datasets, one from Kaggle and one we extracted from a web service called NeoWS which is a RESTful web service from NASA that provides information about near earth asteroids, it updates every day. In overall, the model is tested on two datasets with different features to find the most accurate model to perform the classification., Comment: 6 pages

Quantum computing carries significant potential for addressing practical problems. However, currently available quantum devices suffer from noisy quantum gates, which degrade the fidelity of executed quantum circuits. Therefore, quantum circuit optimization is crucial for obtaining useful results. In this paper, we present QuCLEAR, a compilation framework designed to optimize quantum circuits. QuCLEAR significantly reduces both the two-qubit gate count and the circuit depth through two novel optimization steps. First, we introduce the concept of Clifford Extraction, which extracts Clifford subcircuits to the end of the circuit while optimizing the gates. Second, since Clifford circuits are classically simulatable, we propose Clifford Absorption, which efficiently processes the extracted Clifford subcircuits classically. We demonstrate our framework on quantum simulation circuits, which have wide-ranging applications in quantum chemistry simulation, many-body physics, and combinatorial optimization problems. Near-term algorithms such as VQE and QAOA also fall within this category. Experimental results across various benchmarks show that QuCLEAR achieves up to a $77.7\%$ reduction in CNOT gate count and up to an $84.1\%$ reduction in entangling depth compared to state-of-the-art methods., Comment: 13 pages, 9 figures, 2 tables

NLP-assisted solutions have gained considerable traction to support qualitative data analysis. However, there does not exist a unified evaluation framework that can account for the many different settings in which qualitative researchers may employ them. In this paper, we take a first step in this direction by proposing an evaluation framework to study the way in which different tools may result in different outcomes depending on the collaboration strategy employed. Specifically, we study the impact of synchronous vs. asynchronous collaboration using two different NLP-assisted qualitative research tools and present a comprehensive analysis of significant differences in the consistency, cohesiveness, and correctness of their outputs.

Quantum processing units (QPUs) are highly heterogeneous in terms of physical qubit performance. To add even more complexity, drift in quantum noise landscapes has been well-documented. This makes resource allocation a challenging problem whenever a quantum program must be mapped to hardware. As a solution, we propose a novel resource allocation framework that applies Pauli checks. Pauli checks have demonstrated their efficacy at error mitigation in prior work, and in this paper, we highlight their potential to infer the noise characteristics of a quantum system. Circuits with embedded Pauli checks can be executed on different regions of qubits, and the syndrome data created by error-detecting Pauli checks can be leveraged to guide quantum program outcomes toward regions that produce higher-fidelity final distributions. Using noisy simulation and a real QPU testbed, we show that dynamic quantum resource allocation with Pauli checks can outperform state-of-art mapping techniques, such as those that are noise-aware. Further, when applied toward the Quantum Approximate Optimization Algorithm, techniques guided by Pauli checks demonstrate the ability to increase circuit fidelity 11% on average, and up to 33%.

Hardware accelerators, in particular accelerators for tensor processing, have many potential application domains. However, they currently lack the software infrastructure to support the majority of domains outside of deep learning. Furthermore, a compiler that can easily be updated to reflect changes at both application and hardware levels would enable more agile development and design space exploration of accelerators, allowing hardware designers to realize closer-to-optimal performance. In this work, we discuss how large language models (LLMs) could be leveraged to build such a compiler. Specifically, we demonstrate the ability of GPT-4 to achieve high pass rates in translating code to the Gemmini accelerator, and prototype a technique for decomposing translation into smaller, more LLM-friendly steps. Additionally, we propose a 2-phase workflow for utilizing LLMs to generate hardware-optimized code., Comment: 4 page workshop paper

Matched-filter based gravitational-wave search pipelines identify candidate events within seconds of their arrival on Earth, offering a chance to guide electromagnetic follow-up and observe multi-messenger events. Understanding the detectors' response to an astrophysical transient across the searched signal manifold is paramount to inferring the parameters of the progenitor and deciding which candidates warrant telescope time. We describe a framework that uses artificial neural networks to interpolate gravitational waves and, equivalently, the signal-to noise ratio (SNR) across sufficiently local patches of the signal manifold. Our machine-learning based model generates a single waveform in 6 milliseconds on a CPU and 0.4 milliseconds on a GPU. When using a GPU to generate batches of waveforms simultaneously, we find that we can produce $10^4$ waveforms in $\lesssim 1$ ms. This is achieved while remaining faithful, on average, to 1 part in $10^4$ (1 part in $10^5$) for binary black hole (binary neutron star) waveforms. The model we present is designed to directly utilize intermediate detection pipeline outputs in the hopes of facilitating a better real-time understanding of gravitational-wave candidates.

In this work, we introduce Gemma 2, a new addition to the Gemma family of lightweight, state-of-the-art open models, ranging in scale from 2 billion to 27 billion parameters. In this new version, we apply several known technical modifications to the Transformer architecture, such as interleaving local-global attentions (Beltagy et al., 2020a) and group-query attention (Ainslie et al., 2023). We also train the 2B and 9B models with knowledge distillation (Hinton et al., 2015) instead of next token prediction. The resulting models deliver the best performance for their size, and even offer competitive alternatives to models that are 2-3 times bigger. We release all our models to the community.

We present foundation language models developed to power Apple Intelligence features, including a ~3 billion parameter model designed to run efficiently on devices and a large server-based language model designed for Private Cloud Compute. These models are designed to perform a wide range of tasks efficiently, accurately, and responsibly. This report describes the model architecture, the data used to train the model, the training process, how the models are optimized for inference, and the evaluation results. We highlight our focus on Responsible AI and how the principles are applied throughout the model development.

Formalized in ISO 12913, the "soundscape" approach is a paradigmatic shift towards perception-based urban sound management, aiming to alleviate the substantial socioeconomic costs of noise pollution to advance the United Nations Sustainable Development Goals. Focusing on traffic-exposed outdoor residential sites, we implemented an automatic masker selection system (AMSS) utilizing natural sounds to mask (or augment) traffic soundscapes. We employed a pre-trained AI model to automatically select the optimal masker and adjust its playback level, adapting to changes over time in the ambient environment to maximize "Pleasantness", a perceptual dimension of soundscape quality in ISO 12913. Our validation study involving ($N=68$) residents revealed a significant 14.6 % enhancement in "Pleasantness" after intervention, correlating with increased restorativeness and positive affect. Perceptual enhancements at the traffic-exposed site matched those at a quieter control site with 6 dB(A) lower $L_\text{A,eq}$ and road traffic noise dominance, affirming the efficacy of AMSS as a soundscape intervention, while streamlining the labour-intensive assessment of "Pleasantness" with probabilistic AI prediction., Comment: 41 pages, 4 figures. Preprint submitted to Building and Environment

Reinforcement learning (RL) methods for social robot navigation show great success navigating robots through large crowds of people, but the performance of these learning-based methods tends to degrade in particularly challenging or unfamiliar situations due to the models' dependency on representative training data. To ensure human safety and comfort, it is critical that these algorithms handle uncommon cases appropriately, but the low frequency and wide diversity of such situations present a significant challenge for these data-driven methods. To overcome this challenge, we propose modifications to the learning process that encourage these RL policies to maintain additional caution in unfamiliar situations. Specifically, we improve the Socially Attentive Reinforcement Learning (SARL) policy by (1) modifying the training process to systematically introduce deviations into a pedestrian model, (2) updating the value network to estimate and utilize pedestrian-unpredictability features, and (3) implementing a reward function to learn an effective response to pedestrian unpredictability. Compared to the original SARL policy, our modified policy maintains similar navigation times and path lengths, while reducing the number of collisions by 82% and reducing the proportion of time spent in the pedestrians' personal space by up to 19 percentage points for the most difficult cases. We also describe how to apply these modifications to other RL policies and demonstrate that some key high-level behaviors of our approach transfer to a physical robot.

Background: The high dimensionality of radiomic feature sets, the variability in radiomic feature types and potentially high computational requirements all underscore the need for an effective method to identify the smallest set of predictive features for a given clinical problem. Purpose: Develop a methodology and tools to identify and explain the smallest set of predictive radiomic features. Materials and Methods: 89,714 radiomic features were extracted from five cancer datasets: low-grade glioma, meningioma, non-small cell lung cancer (NSCLC), and two renal cell carcinoma cohorts (n=2104). Features were categorized by computational complexity into morphological, intensity, texture, linear filters, and nonlinear filters. Models were trained and evaluated on each complexity level using the area under the curve (AUC). The most informative features were identified, and their importance was explained. The optimal complexity level and associated most informative features were identified using systematic statistical significance analyses and a false discovery avoidance procedure, respectively. Their predictive importance was explained using a novel tree-based method. Results: MEDimage, a new open-source tool, was developed to facilitate radiomic studies. Morphological features were optimal for MRI-based meningioma (AUC: 0.65) and low-grade glioma (AUC: 0.68). Intensity features were optimal for CECT-based renal cell carcinoma (AUC: 0.82) and CT-based NSCLC (AUC: 0.76). Texture features were optimal for MRI-based renal cell carcinoma (AUC: 0.72). Tuning the Hounsfield unit range improved results for CECT-based renal cell carcinoma (AUC: 0.86). Conclusion: Our proposed methodology and software can estimate the optimal radiomics complexity level for specific medical outcomes, potentially simplifying the use of radiomics in predictive modeling across various contexts.

Leveraging the features of the GstLAL pipeline, we present the results of a matched filtering search for asymmetric binary black hole systems with heavily misaligned spins in LIGO and Virgo data taken during the third observing run. Our target systems show strong imprints of precession whereas current searches have non-optimal sensitivity in detecting them. After measuring the sensitivity improvement brought by our search over standard spin-aligned searches, we report the detection of 30 gravitational wave events already discovered in the latest version of the Gravitational Wave Transient Catalog. However, we do not find any additional significant gravitational wave candidates. Our results allow us to place an upper limit of $R_{90\%} = 0.28^{+0.33}_{-0.04}\;\; \mathrm{Gpc^{-3}yr^{-1}}$ on the merger rate of a hypothetical subpopulation of asymmetric, heavily precessing signals, not identified by other searches. Since our upper limit is consistent with the latest rate estimates from the LIGO-Virgo-KAGRA collaboration, our findings rule out the existence of a yet-to-be-discovered population of precessing binaries.

Memorization in language models is typically treated as a homogenous phenomenon, neglecting the specifics of the memorized data. We instead model memorization as the effect of a set of complex factors that describe each sample and relate it to the model and corpus. To build intuition around these factors, we break memorization down into a taxonomy: recitation of highly duplicated sequences, reconstruction of inherently predictable sequences, and recollection of sequences that are neither. We demonstrate the usefulness of our taxonomy by using it to construct a predictive model for memorization. By analyzing dependencies and inspecting the weights of the predictive model, we find that different factors influence the likelihood of memorization differently depending on the taxonomic category.

Social media personalization algorithms increasingly influence the flow of civic information through society, resulting in concerns about "filter bubbles", "echo chambers", and other ways they might exacerbate ideological segregation and fan the spread of polarizing content. To address these concerns, we designed and conducted a sociotechnical audit (STA) to investigate how Twitter/X's timeline algorithm affects news curation while also tracking how user perceptions change in response. We deployed a custom-built system that, over the course of three weeks, passively tracked all tweets loaded in users' browsers in the first week, then in the second week enacted an intervention to users' Twitter/X homepage to restrict their view to only the algorithmic or chronological timeline (randomized). We flipped this condition for each user in the third week. We ran our audit in late 2023, collecting user-centered metrics (self-reported survey measures) and platform-centered metrics (views, clicks, likes) for 243 users, along with over 800,000 tweets. Using the STA framework, our results are two-fold: (1) Our algorithm audit finds that Twitter/X's algorithmic timeline resulted in a lower quantity but higher quality of news -- less ideologically congruent, less extreme, and slightly more reliable -- compared to the chronological timeline. (2) Our user audit suggests that although our timeline intervention had significant effects on users' behaviors, it had little impact on their overall perceptions of the platform. Our paper discusses these findings and their broader implications in the context of algorithmic news curation, user-centric audits, and avenues for independent social science research., Comment: 24 pages, 5 figures, Computer-Supported Cooperative Work

Pauli Check Sandwiching (PCS) is an error mitigation scheme that uses pairs of parity checks to detect errors in the payload circuit. While increasing the number of check pairs improves error detection, it also introduces additional noise to the circuit and exponentially increases the required sampling size. To address these limitations, we propose a novel error mitigation scheme, Pauli Check Extrapolation (PCE), which integrates PCS with an extrapolation technique similar to Zero-Noise Extrapolation (ZNE). However, instead of extrapolating to the `zero-noise' limit, as is done in ZNE, PCE extrapolates to the `maximum check' limit--the number of check pairs theoretically required to achieve unit fidelity. In this study, we focus on applying a linear model for extrapolation and also derive a more general exponential ansatz based on the Markovian error model. We demonstrate the effectiveness of PCE by using it to mitigate errors in the shadow estimation protocol, particularly for states prepared by the variational quantum eigensolver (VQE). Our results show that this method can achieve higher fidelities than the state-of-the-art Robust Shadow (RS) estimation scheme, while significantly reducing the number of required samples by eliminating the need for a calibration procedure. We validate these findings on both fully-connected topologies and simulated IBM hardware backends.

Programming models for distributed dataflow have long focused on analytical workloads that allow the runtime to dynamically place and schedule compute logic. Meanwhile, models that enable fine-grained control over placement, such as actors, make global optimization difficult. In this extended abstract, we present Suki, an embedded Rust DSL that lets developers implement streaming dataflow with explicit placement of computation. Key to this choreographic programming approach is our use of staged programming, which lets us expose a high-level Rust API while compiling local compute units into individual binaries with zero-overhead. We also explore how this approach, combined with Rust's trait system, enables a type-safe API for mapping dataflow programs to cloud computing resources.

Reducing the inference latency of large language models (LLMs) is crucial, and speculative decoding (SD) stands out as one of the most effective techniques. Rather than letting the LLM generate all tokens directly, speculative decoding employs effective proxies to predict potential outputs, which are then verified by the LLM without compromising the generation quality. Yet, deploying SD in real online LLM serving systems (with continuous batching) does not always yield improvement -- under higher request rates or low speculation accuracy, it paradoxically increases latency. Furthermore, there is no best speculation length work for all workloads under different system loads. Based on the observations, we develop a dynamic framework SmartSpec. SmartSpec dynamically determines the best speculation length for each request (from 0, i.e., no speculation, to many tokens) -- hence the associated speculative execution costs -- based on a new metric called goodput, which characterizes the current observed load of the entire system and the speculation accuracy. We show that SmartSpec consistently reduces average request latency by up to 3.2x compared to non-speculative decoding baselines across different sizes of target models, draft models, request rates, and datasets. Moreover, SmartSpec can be applied to different styles of speculative decoding, including traditional, model-based approaches as well as model-free methods like prompt lookup and tree-style decoding.

Human children far exceed modern machine learning algorithms in their sample efficiency, achieving high performance in key domains with much less data than current models. This ''data gap'' is a key challenge both for building intelligent artificial systems and for understanding human development. Egocentric video capturing children's experience -- their ''training data'' -- is a key ingredient for comparison of humans and models and for the development of algorithmic innovations to bridge this gap. Yet there are few such datasets available, and extant data are low-resolution, have limited metadata, and importantly, represent only a small set of children's experiences. Here, we provide the first release of the largest developmental egocentric video dataset to date -- the BabyView dataset -- recorded using a high-resolution camera with a large vertical field-of-view and gyroscope/accelerometer data. This 493 hour dataset includes egocentric videos from children spanning 6 months - 5 years of age in both longitudinal, at-home contexts and in a preschool environment. We provide gold-standard annotations for the evaluation of speech transcription, speaker diarization, and human pose estimation, and evaluate models in each of these domains. We train self-supervised language and vision models and evaluate their transfer to out-of-distribution tasks including syntactic structure learning, object recognition, depth estimation, and image segmentation. Although performance in each scales with dataset size, overall performance is relatively lower than when models are trained on curated datasets, especially in the visual domain. Our dataset stands as an open challenge for robust, humanlike AI systems: how can such systems achieve human-levels of success on the same scale and distribution of training data as humans?, Comment: 9 pages, 2 figures, 4 tables and SI. Submitted to NeurIPS Datasets and Benchmarks

How (dis)similar are the learning trajectories of vision-language models and children? Recent modeling work has attempted to understand the gap between models' and humans' data efficiency by constructing models trained on less data, especially multimodal naturalistic data. However, such models are often evaluated on adult-level benchmarks, with limited breadth in language abilities tested, and without direct comparison to behavioral data. We introduce DevBench, a multimodal benchmark comprising seven language evaluation tasks spanning the domains of lexical, syntactic, and semantic ability, with behavioral data from both children and adults. We evaluate a set of vision-language models on these tasks, comparing models and humans not only on accuracy but on their response patterns. Across tasks, models exhibit variation in their closeness to human response patterns, and models that perform better on a task also more closely resemble human behavioral responses. We also examine the developmental trajectory of OpenCLIP over training, finding that greater training results in closer approximations to adult response patterns. DevBench thus provides a benchmark for comparing models to human language development. These comparisons highlight ways in which model and human language learning processes diverge, providing insight into entry points for improving language models., Comment: Accepted at NeurIPS 2024 (Oral)