Search

We present 450$\mu$m and 850$\mu$m James Clerk Maxwell Telescope (JCMT) observations of the Corona Australis (CrA) molecular cloud taken as part of the JCMT Gould Belt Legacy Survey (GBLS). We present a catalogue of 39 starless and protostellar sources, for which we determine source temperatures and masses using SCUBA-2 450$\mu$m/850$\mu$m flux density ratios for sources with reliable 450$\mu$m detections, and compare these to values determined using temperatures measured by the Herschel Gould Belt Survey (HGBS). In keeping with previous studies, we find that SCUBA-2 preferentially detects high-volume-density starless cores, which are most likely to be prestellar (gravitationally bound). We do not observe any anti-correlation between temperature and volume density in the starless cores in our sample. Finally, we combine our SCUBA-2 and Herschel data to perform SED fitting from 160-850$\mu$m across the central Coronet region, thereby measuring dust temperature $T$, dust emissivity index $\beta$ and column density $N({\rm H}_2)$ across the Coronet. We find that $\beta$ varies across the Coronet, particularly measuring $\beta = 1.55 \pm 0.35$ in the colder starless SMM-6 clump to the north of the B star R CrA. This relatively low value of $\beta$ is suggestive of the presence of large dust grains in SMM-6, even when considering the effects of $T-\beta$ fitting degeneracy and $^{12}$CO contamination of SCUBA-2 850$\mu$m data on the measured $\beta$ values., Comment: 27 pages, 18 figures, 4 tables. Accepted for publication in MNRAS more...

We present an approach for acquiring grounded representations of words from mixed-initiative, situated interactions with a human instructor. The work focuses on the acquisition of diverse types of knowledge including perceptual, semantic, and procedural knowledge along with learning grounded meanings. Interactive learning allows the agent to control its learning by requesting instructions about unknown concepts, making learning efficient. Our approach has been instantiated in Soar and has been evaluated on a table-top robotic arm capable of manipulating small objects. more...

The atmospheres of planets between the size of Earth and Neptune at short orbital periods have been under intense scrutiny. Of the ~dozen planets in this regime with atmospheres studied so far, a few appear to have prominent molecular features while others appear relatively void of detectable atmospheres. Further work is therefore needed to understand the atmospheres of these planets, starting with observing a larger sample. To this end, we present the 3-5 micron transmission spectrum of TOI-776 c, a warm (Teq ~420 K), ~2 Rearth, ~7 Mearth planet orbiting an M1V star, measured with JWST NIRSpec/G395H. By combining two visits, we measure a median transit precision of ~18 ppm and ~32 ppm in the NRS1 and NRS2 detectors, respectively. We compare the transmission spectrum to both non-physical and physical models, and find no strong evidence for molecular features. For cloud-top pressures larger than 10^-3 bar, we rule out atmospheric metallicities less than 180-240x solar (depending on the reduction and modeling technique), which corresponds to a mean molecular weight of ~6-8 g/mol. However, we find simple atmosphere mixture models (H2O+H2/He or CO2+H2/He) give more pessimistic constraints, and caution that mean molecular weight inferences are model dependent. We compare TOI-776 c to the similar planet TOI-270 d, and discuss possible options for further constraining TOI-776 c's atmospheric composition. Overall, we suggest these TOI-776 c observations may represent a combination of planetary and stellar parameters that fall just below the threshold of detectable features in small planet spectra; finding this boundary is one of the main goals of the COMPASS program., Comment: 21 pages, 15 figures, 4 tables. Accepted for publication in AJ more...

Recent research across mathematical problem solving, proof assistant programming and multimodal jailbreaking documents a striking finding: when (multimodal) language model tackle a suite of tasks with multiple attempts per task -- succeeding if any attempt is correct -- then the negative log of the average success rate scales a power law in the number of attempts. In this work, we identify an apparent puzzle: a simple mathematical calculation predicts that on each problem, the failure rate should fall exponentially with the number of attempts. We confirm this prediction empirically, raising a question: from where does aggregate polynomial scaling emerge? We then answer this question by demonstrating per-problem exponential scaling can be made consistent with aggregate polynomial scaling if the distribution of single-attempt success probabilities is heavy tailed such that a small fraction of tasks with extremely low success probabilities collectively warp the aggregate success trend into a power law - even as each problem scales exponentially on its own. We further demonstrate that this distributional perspective explains previously observed deviations from power law scaling, and provides a simple method for forecasting the power law exponent with an order of magnitude lower relative error, or equivalently, ${\sim}2-4$ orders of magnitude less inference compute. Overall, our work contributes to a better understanding of how neural language model performance improves with scaling inference compute and the development of scaling-predictable evaluations of (multimodal) language models. more...

Interpreting data is central to modern research. Large language models (LLMs) show promise in providing such natural language interpretations of data, yet simple feature extraction methods such as prompting often fail to produce accurate and versatile descriptions for diverse datasets and lack control over granularity and scale. To address these limitations, we propose a domain-agnostic method for dataset featurization that provides precise control over the number of features extracted while maintaining compact and descriptive representations comparable to human expert labeling. Our method optimizes the selection of informative binary features by evaluating the ability of an LLM to reconstruct the original data using those features. We demonstrate its effectiveness in dataset modeling tasks and through two case studies: (1) Constructing a feature representation of jailbreak tactics that compactly captures both the effectiveness and diversity of a larger set of human-crafted attacks; and (2) automating the discovery of features that align with human preferences, achieving accuracy and robustness comparable to expert-crafted features. Moreover, we show that the pipeline scales effectively, improving as additional features are sampled, making it suitable for large and diverse datasets. more...

Large Language Models (LLMs) are becoming increasingly capable across global languages. However, the ability to communicate across languages does not necessarily translate to appropriate cultural representations. A key concern is US-centric bias, where LLMs reflect US rather than local cultural values. We propose a novel methodology that compares LLM-generated response distributions against population-level opinion data from the World Value Survey across four languages (Danish, Dutch, English, and Portuguese). Using a rigorous linear mixed-effects regression framework, we compare two families of models: Google's Gemma models (2B--27B parameters) and successive iterations of OpenAI's turbo-series. Across the families of models, we find no consistent relationships between language capabilities and cultural alignment. While the Gemma models have a positive correlation between language capability and cultural alignment across languages, the OpenAI models do not. Importantly, we find that self-consistency is a stronger predictor of multicultural alignment than multilingual capabilities. Our results demonstrate that achieving meaningful cultural alignment requires dedicated effort beyond improving general language capabilities. more...

LLM developers have imposed technical interventions to prevent fine-tuning misuse attacks, attacks where adversaries evade safeguards by fine-tuning the model using a public API. Previous work has established several successful attacks against specific fine-tuning API defences. In this work, we show that defences of fine-tuning APIs that seek to detect individual harmful training or inference samples ('pointwise' detection) are fundamentally limited in their ability to prevent fine-tuning attacks. We construct 'pointwise-undetectable' attacks that repurpose entropy in benign model outputs (e.g. semantic or syntactic variations) to covertly transmit dangerous knowledge. Our attacks are composed solely of unsuspicious benign samples that can be collected from the model before fine-tuning, meaning training and inference samples are all individually benign and low-perplexity. We test our attacks against the OpenAI fine-tuning API, finding they succeed in eliciting answers to harmful multiple-choice questions, and that they evade an enhanced monitoring system we design that successfully detects other fine-tuning attacks. We encourage the community to develop defences that tackle the fundamental limitations we uncover in pointwise fine-tuning API defences. more...

Automatic evaluation methods based on large language models (LLMs) are emerging as the standard tool for assessing the instruction-following abilities of LLM-based agents. The most common method in this paradigm, pairwise comparisons with a baseline model, critically depends on the assumption of transitive preferences. However, the validity of this assumption remains largely unexplored. In this study, we investigate the presence of non-transitivity within the AlpacaEval framework and analyze its effects on model rankings. We find that LLM judges exhibit non-transitive preferences, leading to rankings that are sensitive to the choice of the baseline model. To mitigate this issue, we show that round-robin tournaments combined with Bradley-Terry models of preference can produce more reliable rankings. Notably, our method increases both the Spearman correlation and the Kendall correlation with Chatbot Arena (95.0% -> 96.4% and 82.1% -> 86.3% respectively). To address the computational cost of round-robin tournaments, we propose Swiss-Wise Iterative Matchmaking (Swim) tournaments, using a dynamic matching strategy to capture the benefits of round-robin tournaments while maintaining computational efficiency., Comment: 8 pages, 6 figures, 2 tables (30 pages, 11 figures, 8 tables including references and appendices) more...

We present a federated learning approach for Bayesian model-based clustering of large-scale binary and categorical datasets. We introduce a principled 'divide and conquer' inference procedure using variational inference with local merge and delete moves within batches of the data in parallel, followed by 'global' merge moves across batches to find global clustering structures. We show that these merge moves require only summaries of the data in each batch, enabling federated learning across local nodes without requiring the full dataset to be shared. Empirical results on simulated and benchmark datasets demonstrate that our method performs well in comparison to existing clustering algorithms. We validate the practical utility of the method by applying it to large scale electronic health record (EHR) data. more...

We report direct measurement of the dispersion relation of polaritons in GaAs/AlGaAs microcavity structures at room temperature, which clearly shows that the polaritons are in the strong coupling limit. The Rabi splitting of the polariton states decreases as the polariton gas increases in density, but even when the polariton gas becomes a coherent, Bose-condensate-like state, the polaritons retain a strong exciton component, as seen in the nonlinear energy shift of the light emission. This opens up the possibility of polaritonic devices at room temperature in a material system which can be grown with very high quality and uniformity. more...

The Transiting Exoplanet Survey Satellite (TESS) mission identified a potential 0.88 REarth planet with a period of 7.577 days, orbiting the nearby M1V star GJ 341 (TOI 741.01). This system has already been observed by the James Webb Space Telescope (JWST) to search for presence of an atmosphere on this planet. Here, we present an in-depth analysis of the GJ 341 system using all available public data. We provide improved parameters for the host star, an updated value of the planet radius, and support the planetary nature of the object (now GJ 341 b). We use 57 HARPS radial velocities to model the magnetic cycle and activity of the host star, and constrain the mass of GJ 341 b to upper limits of 4.0 MEarth (3 sigma) and 2.9 MEarth (1 sigma). We also rule out the presence of additional companions with M sin i > 15.1 MEarth, and P < 1750 days, and the presence of contaminating background objects during the TESS and JWST observations. These results provide key information to aid the interpretation of the recent JWST atmospheric observations and other future observations of this planet., Comment: 18 pages, 17 figures more...

Many questions in quantitative finance, uncertainty quantification, and other disciplines are answered by computing the population mean, $\mu := \mathbb{E}(Y)$, where instances of $Y:=f(\boldsymbol{X})$ may be generated by numerical simulation and $\boldsymbol{X}$ has a simple probability distribution. The population mean can be approximated by the sample mean, $\hat{\mu}_n := n^{-1} \sum_{i=0}^{n-1} f(\boldsymbol{x}_i)$ for a well chosen sequence of nodes, $\{\boldsymbol{x}_0, \boldsymbol{x}_1, \ldots\}$ and a sufficiently large sample size, $n$. Computing $\mu$ is equivalent to computing a $d$-dimensional integral, $\int f(\boldsymbol{x}) \varrho(\boldsymbol{x}) \, \mathrm{d} \boldsymbol{x}$, where $\varrho$ is the probability density for $\boldsymbol{X}$. Quasi-Monte Carlo methods replace independent and identically distributed sequences of random vector nodes, $\{\boldsymbol{x}_i \}_{i = 0}^{\infty}$, by low discrepancy sequences. This accelerates the convergence of $\hat{\mu}_n$ to $\mu$ as $n \to \infty$. This tutorial describes low discrepancy sequences and their quality measures. We demonstrate the performance gains possible with quasi-Monte Carlo methods. Moreover, we describe how to formulate problems to realize the greatest performance gains using quasi-Monte Carlo. We also briefly describe the use of quasi-Monte Carlo methods for problems beyond computing the mean, $\mu$. more...

Humans strive to design safe AI systems that align with our goals and remain under our control. However, as AI capabilities advance, we face a new challenge: the emergence of deeper, more persistent relationships between humans and AI systems. We explore how increasingly capable AI agents may generate the perception of deeper relationships with users, especially as AI becomes more personalised and agentic. This shift, from transactional interaction to ongoing sustained social engagement with AI, necessitates a new focus on socioaffective alignment-how an AI system behaves within the social and psychological ecosystem co-created with its user, where preferences and perceptions evolve through mutual influence. Addressing these dynamics involves resolving key intrapersonal dilemmas, including balancing immediate versus long-term well-being, protecting autonomy, and managing AI companionship alongside the desire to preserve human social bonds. By framing these challenges through a notion of basic psychological needs, we seek AI systems that support, rather than exploit, our fundamental nature as social and emotional beings. more...

Evaluations of large language model (LLM) risks and capabilities are increasingly being incorporated into AI risk management and governance frameworks. Currently, most risk evaluations are conducted by designing inputs that elicit harmful behaviors from the system. However, a fundamental limitation of this approach is that the harmfulness of the behaviors identified during any particular evaluation can only lower bound the model's worst-possible-case behavior. As a complementary method for eliciting harmful behaviors, we propose evaluating LLMs with model tampering attacks which allow for modifications to latent activations or weights. We pit state-of-the-art techniques for removing harmful LLM capabilities against a suite of 5 input-space and 6 model tampering attacks. In addition to benchmarking these methods against each other, we show that (1) model resilience to capability elicitation attacks lies on a low-dimensional robustness subspace; (2) the attack success rate of model tampering attacks can empirically predict and offer conservative estimates for the success of held-out input-space attacks; and (3) state-of-the-art unlearning methods can easily be undone within 16 steps of fine-tuning. Together these results highlight the difficulty of removing harmful LLM capabilities and show that model tampering attacks enable substantially more rigorous evaluations than input-space attacks alone. We release models at https://huggingface.co/LLM-GAT more...

The next generation of photonic circuits will require programmable, ultrafast, and energy-efficient components on a scalable platform for quantum and neuromorphic computing. Here, we present ultrafast electrical control of highly nonlinear light-matter hybrid quasi-particles, called waveguide exciton-dipolaritons, with extremely low power consumption. Our device performs as an optical transistor with a GHz-rate electrical modulation at a record-low total energy consumption $\sim$3 fJ/bit and a compact active area of down to 25 $\mu$m$^2$. This work establishes waveguide-dipolariton platforms for scalable, electrically reconfigurable, ultra-low power photonic circuits for both classical and quantum computing and communication., Comment: 7 pages, 3 figures, 1 table, Supplementary materials more...

We present two transit observations of the $\sim$520K, 1.85R$_\oplus$, 4.0M$_\oplus$ super-Earth TOI-776b with JWST NIRSpec/G395H, resulting in a 2.8-5.2$\mu$m transmission spectrum. Producing reductions using the ExoTiC-JEDI and Eureka! pipelines, we obtain a median transit depth precision of 34ppm for both visits and both reductions in spectroscopic channels 30 pixels wide ($\sim$0.02$\mu$m). We find that our independent reductions produce consistent transmission spectra, however, each visit shows differing overall structure. For both reductions, a flat line is preferred for Visit 1 while a flat line with an offset between the NRS1 and NRS2 detectors is preferred for Visit 2; however, we are able to correct for this offset during our modeling analysis following methods outlined in previous literature. Using picaso forward models, we can rule out metallicities up to at least 100$\times$ solar with an opaque pressure of 10$^{-3}$ bar to $\geq$3$\sigma$ in all cases, however, the exact lower limit varies between the visits, with Visit 1 ruling out $\lesssim$100$\times$ solar while the lower limits for Visit 2 extend beyond $\sim$350$\times$ solar. Our results add to the growing list of super-Earth atmospheric constraints by JWST, which provide critical insight into the diversity and challenges of characterizing terrestrial planets., Comment: 17 pages, 7 figures, 4 tables. Accepted for publication in AJ. arXiv admin note: text overlap with arXiv:2404.00093 more...

Coronal holes (CH) are magnetically open regions that allow hot coronal plasma to escape from the Sun and form the high-speed solar wind. This wind can interact with Earth's magnetic field. For this reason, developing an accurate understanding of CH regions is vital for understanding space weather and its effects on Earth. The process of identifying CH regions typically relies on extreme ultraviolet (EUV) imagery, leveraging the fact that CHs appear dark at these wavelengths. Accurate identification of CHs in EUV, however, can be difficult due to a variety of factors, including stray light from nearby regions, limb brightening, and the presence of filaments (which also appear dark, but are not sources of solar wind). In order to overcome these issues, this work incorporates photospheric magnetic field data into a classical EUV-based segmentation algorithm based on the active contours without edges (ACWE) segmentation method. In this work magnetic field data are incorporated directly into the segmentation process, serving both as a method for removing non-CH regions in advance, and as a method to constrain evolution of the boundary which identifies the CH boundary. This reduces the presence of filaments while simultaneously allowing the segmentation to include CH regions that may be difficult to identify due to the aforementioned inconsistent intensities. more...

We present the first analytic-derivative-based formulation of vibrational circular dichroism (VCD) atomic axial tensors for second-order Moller-Plesset (MP2) perturbation theory. We compare our implementation to our recently reported finite-difference approach and find close agreement, thus validating the new formulation. The new approach is dramatically less computationally expensive than the numerical-derivative method with an overall computational scaling of $O(N^6)$. In addition, we report the first fully analytic VCD spectrum for (S)-methyloxirane at the MP2 level of theory., Comment: 23 pages, 3 tables, 2 figures more...

This paper presents a pipeline for mitigating gender bias in large language models (LLMs) used in medical literature by neutralizing gendered occupational pronouns. A dataset of 379,000 PubMed abstracts from 1965-1980 was processed to identify and modify pronouns tied to professions. We developed a BERT-based model, ``Modern Occupational Bias Elimination with Refined Training,'' or ``MOBERT,'' trained on these neutralized abstracts, and compared its performance with ``1965Bert,'' trained on the original dataset. MOBERT achieved a 70\% inclusive replacement rate, while 1965Bert reached only 4\%. A further analysis of MOBERT revealed that pronoun replacement accuracy correlated with the frequency of occupational terms in the training data. We propose expanding the dataset and refining the pipeline to improve performance and ensure more equitable language modeling in medical applications., Comment: 9 pages, 4 figures more...

As AI systems become more capable, widely deployed, and increasingly autonomous in critical areas such as cybersecurity, biological research, and healthcare, ensuring their safety and alignment with human values is paramount. Machine unlearning -- the ability to selectively forget or suppress specific types of knowledge -- has shown promise for privacy and data removal tasks, which has been the primary focus of existing research. More recently, its potential application to AI safety has gained attention. In this paper, we identify key limitations that prevent unlearning from serving as a comprehensive solution for AI safety, particularly in managing dual-use knowledge in sensitive domains like cybersecurity and chemical, biological, radiological, and nuclear (CBRN) safety. In these contexts, information can be both beneficial and harmful, and models may combine seemingly harmless information for harmful purposes -- unlearning this information could strongly affect beneficial uses. We provide an overview of inherent constraints and open problems, including the broader side effects of unlearning dangerous knowledge, as well as previously unexplored tensions between unlearning and existing safety mechanisms. Finally, we investigate challenges related to evaluation, robustness, and the preservation of safety features during unlearning. By mapping these limitations and open challenges, we aim to guide future research toward realistic applications of unlearning within a broader AI safety framework, acknowledging its limitations and highlighting areas where alternative approaches may be required. more...

Over the past decade, observations of evaporating exoplanets have become increasingly common, driven by the discovery of the near-infrared helium-triplet line as a powerful probe of atmospheric escape. This process significantly influences the evolution of exoplanets, particularly those smaller than Jupiter. Both theoretical and observational studies have aimed to determine how efficiently exoplanets convert their host star's X-ray and ultraviolet (XUV) radiation into atmospheric mass loss. In this study, we employ the open-source atmospheric escape model p-winds to systematically analyze all publicly available helium triplet spectroscopic detections related to exoplanetary atmospheric escape. Our findings indicate that the retrieved outflows strongly depend on the ratio of XUV flux to planetary density ($F_{\text{XUV}}/\rho_p$), supporting the theoretical framework of energy-limited mass loss. We constrain population-level photoevaporative efficiencies to $0.34 \pm 0.13$ and $0.75 \pm 0.21$ for hydrogen-helium fractions of $0.90$ and $0.99$, respectively. These results offer new insights into exoplanetary atmospheric evolution and will aid future studies on exoplanet population demographics., Comment: 18 pages, 4 figures, accepted to ApJ more...

Background: Healthcare has many manual processes that can benefit from automation and augmentation with Generative Artificial Intelligence (AI), the medical billing and coding process. However, current foundational Large Language Models (LLMs) perform poorly when tasked with generating accurate International Classification of Diseases, 10th edition, Clinical Modification (ICD-10-CM) and Current Procedural Terminology (CPT) codes. Additionally, there are many security and financial challenges in the application of generative AI to healthcare. We present a strategy for developing generative AI tools in healthcare, specifically for medical billing and coding, that balances accuracy, accessibility, and patient privacy. Methods: We fine tune the PHI-3 Mini and PHI-3 Medium LLMs using institutional data and compare the results against the PHI-3 base model, a PHI-3 RAG application, and GPT-4o. We use the post operative surgical report as input and the patients billing claim the associated ICD-10, CPT, and Modifier codes as the target result. Performance is measured by accuracy of code generation, proportion of invalid codes, and the fidelity of the billing claim format. Results: Both fine-tuned models performed better or as well as GPT-4o. The Phi-3 Medium fine-tuned model showed the best performance (ICD-10 Recall and Precision: 72%, 72%; CPT Recall and Precision: 77%, 79%; Modifier Recall and Precision: 63%, 64%). The Phi-3 Medium fine-tuned model only fabricated 1% of ICD-10 codes and 0.6% of CPT codes generated. Conclusions: Our study shows that a small model that is fine-tuned on domain-specific data for specific tasks using a simple set of open-source tools and minimal technological and monetary requirements performs as well as the larger contemporary consumer models., Comment: 21 pages, 3 figures, 2 tables more...

Electrocatalytic processes occurring at the heterogeneous interface are complex and their understanding at the molecular level remains challenging. Atomic force microscope (AFM) can detect force interactions down to the atomic level, but so far it has been mainly used to obtain in-situ images of electrocatalysts. Here for the first time, we employ AFM to investigate gas evolution at a platinum ultramicroelectrode (Pt UME) under electrochemical conditions using the force noise method. We detect excess force noise when the individual H2 gas bubble nucleation, growth, and detachment events occur at the Pt UME. The excess noise varies linearly with the applied potential on a semi-log plot. Chronoamperometry current fluctuations indicate that the H2 gas bubble radius increases from 1 to 10 micrometers and the AFM deflection signal measurements indicate that the mean H2 bubble radius is 321 micrometers. These two values point to the coalescence of small bubbles generated at the Pt UME interface. The contribution reports an innovative method to probe the gas bubble dynamics at the heterogenous interface under electrochemical conditions with high sensitivity. more...

Efficiently estimating large numbers of non-commuting observables is an important subroutine of many quantum science tasks. We present the derandomized shallow shadows (DSS) algorithm for efficiently learning a large set of non-commuting observables, using shallow circuits to rotate into measurement bases. Exploiting tensor network techniques to ensure polynomial scaling of classical resources, our algorithm outputs a set of shallow measurement circuits that approximately minimizes the sample complexity of estimating a given set of Pauli strings. We numerically demonstrate systematic improvement, in comparison with state-of-the-art techniques, for energy estimation of quantum chemistry benchmarks and verification of quantum many-body systems, and we observe DSS's performance consistently improves as one allows deeper measurement circuits. These results indicate that in addition to being an efficient, low-depth, stand-alone algorithm, DSS can also benefit many larger quantum algorithms requiring estimation of multiple non-commuting observables., Comment: 10+29 pages, 9 figures more...

As part of the BISTRO survey, we present JCMT 850 $\mu$m polarimetric observations towards the Orion Integral-Shaped Filament (ISF) that covers three portions known as OMC-1, OMC-2, and OMC-3. The magnetic field threading the ISF seen in the JCMT POL-2 map appears as a tale of three: pinched for OMC-1, twisted for OMC-2, and nearly uniform for OMC-3. A multi-scale analysis shows that the magnetic field structure in OMC-3 is very consistent at all the scales, whereas the field structure in OMC-2 shows no correlation across different scales. In OMC-1, the field retains its mean orientation from large to small scales, but shows some deviations at small scales. Histograms of relative orientations between the magnetic field and filaments reveal a bimodal distribution for OMC-1, a relatively random distribution for OMC-2, and a distribution with a predominant peak at 90$^\circ$ for OMC-3. Furthermore, the magnetic fields in OMC-1 and OMC-3 both appear to be aligned perpendicular to the fibers, which are denser structures within the filament, but the field in OMC-2 is aligned along with the fibers. All these suggest that gravity, turbulence, and magnetic field are each playing a leading role in OMC-1, 2, and 3, respectively. While OMC-2 and 3 have almost the same gas mass, density, and non-thermal velocity dispersion, there are on average younger and fewer young stellar objects in OMC-3, providing evidence that a stronger magnetic field will induce slower and less efficient star formation in molecular clouds., Comment: published in the ApJ Letters more...

Krylov quantum diagonalization methods for ground state energy estimation have emerged as a compelling use case for quantum computers. However, many existing methods rely on subroutines, in particular the Hadamard test, that are challenging on near-term quantum hardware. Motivated by this problem, we present a quantum Krylov method that uses only time evolutions and recovery probabilities, making it well adapted for current quantum computers. This is supplemented with a classical post-processing derivative estimation algorithm. The method ultimately estimates the eigenvalues of the commutator super-operator $X\to[H,X]$, so we declare it a super-Krylov method. We propose applying this method to estimate the ground-state energy of two classes of Hamiltonians: where either the highest energy is easily computable, or where the lowest and highest energies have the same absolute value. We prove that the resulting ground energy estimate converges in the noise-free regime and provide a classical numerical demonstration of the method in the presence of noise. more...

Coupled many-body quantum systems give rise to rich emergent physics and abundance of both stationary and dynamical behaviours. Designing platforms with tunable and distinct forms of coupling gives new insight into the collective behaviour of the dimerised quantum systems. Fundamentally, two systems can exchange particles through either forbidden or allowed channels, underpinning evanescent and ballistic coupling mechanisms, respectively. Based on proximity, the former leads to large spectral splitting that defines, for instance, chemical binding energies, whereas the latter pushes for stringent phase-matching and synchronicity between oscillating degrees-of-freedom, analogous to phase-coupled harmonic oscillators, with a significantly smaller impact on the energy landscape. Here, we demonstrate an all-optically tunable evanescent-ballistic quantum fluid dimer based on hyperbolic exciton-polariton condensates in a photonic crystal waveguide. By changing the angle of the polariton dimer relative to the grating, the system transitions from an evanescently-coupled molecule with large mode-splitting to a ballistic condensate dimer with strict phase-matching conditions and rich interference patterns. We directly measure the condensate spectral features and mass flow subject to a saddle dispersion relation, and connect our results to mean field theory. These results showcase the potential of photonic crystals to study condensed matter phenomena lying at the interface between delay-coupled nonlinear oscillators and tight binding physics. more...

We present a compact quantum circuit for factoring a large class of integers, including some whose classical hardness is expected to be equivalent to RSA (but not including RSA integers themselves). To our knowledge, it is the first polynomial-time circuit to achieve sublinear qubit count for a classically-hard factoring problem; the circuit also achieves sublinear depth and nearly linear gate count. We build on the quantum algorithm for squarefree decomposition discovered by Li, Peng, Du and Suter (Nature Scientific Reports 2012), which relies on computing the Jacobi symbol in quantum superposition. Our circuit completely factors any number $N$, whose prime decomposition has distinct exponents, and finds at least one non-trivial factor if not all exponents are the same. In particular, to factor an $n$-bit integer $N=P^2 Q$ (with $P$ and $Q$ prime, and $Q<2^m$ for some $m$), our circuit uses $\tilde{O}(m)$ qubits and has depth at most $\tilde{O}(m + n/m)$, with $\tilde{O}(n)$ quantum gates. When $m=\Theta(n^a)$ with $2/3 < a < 1$, the space and depth are sublinear in $n$, yet no known classical algorithms exploit the relatively small size of $Q$ to run faster than general-purpose factoring algorithms. We thus believe that factoring such numbers has potential to be the most concretely efficient classically-verifiable proof of quantumness currently known. The technical core of our contribution is a new space-efficient and parallelizable quantum algorithm to compute the Jacobi symbol of $A$ mod $B$, in the regime where $B$ is classical and much larger than $A$. In the context of the larger Jacobi algorithm for factoring $N = P^2Q$, this reduces the overall qubit count to be roughly proportional to the length of $Q$, rather than the length of $N$. Finally, we note that our circuit for computing the Jacobi symbol generalizes to related problems, such as computing the GCD. more...

We consider laminar forced convection in a shrouded longitudinal-fin heat sink (LFHS) with tip clearance, as described by the pioneering study of Sparrow, Baliga and Patankar [1978, J. Heat Trans, 100(4)]. The base of the LFHS is isothermal but the fins, while thin, are not isothermal, i.e., the conjugate heat transfer problem is of interest. Whereas Sparrow et al. solved the fully developed flow and thermal problems numerically for a range of geometries and fin conductivities, we consider here the physically realistic asymptotic limit where the fins are closely spaced, i.e. the spacing is small relative to their height and the clearance above them. The flow problem in this limit was considered by Miyoshi et al. [2024, J. Fluid Mech, 991, A2], and here we consider the corresponding thermal problem. Using matched asymptotic expansions, we find explicit solutions for the temperature field (in both the fluid and fins) and conjugate Nusselt numbers (local and average). The structure of the asymptotic solutions provides insight into the results of Sparrow et al.: the flow is highest in the gap above the fins, hence heat transfer predominantly occurs close to the fin tips. The new formulas are compared to numerical solutions and are found to be accurate for practical LFHSs. Significantly, existing analytical results for ducts are for boundaries that are either wholly isothermal, wholly isoflux or with one of these conditions on each wall. Consequently, this study provides the first analytical results for conjugate Nusselt numbers for flow through ducts., Comment: 39 pages. Supplementary material in ancillary files more...

Biclustering has gained interest in gene expression data analysis due to its ability to identify groups of samples that exhibit similar behaviour in specific subsets of genes (or vice versa), in contrast to traditional clustering methods that classify samples based on all genes. Despite advances, biclustering remains a challenging problem, even with cutting-edge methodologies. This paper introduces an extension of the recently proposed Spike-and-Slab Lasso Biclustering (SSLB) algorithm, termed Outcome-Guided SSLB (OG-SSLB), aimed at enhancing the identification of biclusters in gene expression analysis. Our proposed approach integrates disease outcomes into the biclustering framework through Bayesian profile regression. By leveraging additional clinical information, OG-SSLB improves the interpretability and relevance of the resulting biclusters. Comprehensive simulations and numerical experiments demonstrate that OG-SSLB achieves superior performance, with improved accuracy in estimating the number of clusters and higher consensus scores compared to the original SSLB method. Furthermore, OG-SSLB effectively identifies meaningful patterns and associations between gene expression profiles and disease states. These promising results demonstrate the effectiveness of OG-SSLB in advancing biclustering techniques, providing a powerful tool for uncovering biologically relevant insights. The OGSSLB software can be found as an R/C++ package at https://github.com/luisvargasmieles/OGSSLB . more...

Context: The study of quasar outflows is essential in understanding the connection between active galactic nuclei (AGN) and their host galaxies. We analyze the VLT/UVES spectrum of quasar SDSS J0932+0840 and identify several narrow and broad outflow components in absorption, with multiple ionization species including Fe II, which puts it among a rare class of outflows known as FeLoBALs. Aims: We study one of the outflow components to determine its physical characteristics by determining the total hydrogen column density, ionization parameter and the hydrogen number density. Through these parameters, we aim to obtain the distance of the outflow from the central source, its mass outflow rate and kinetic luminosity, and to constrain the contribution of the outflow to AGN feedback. Methods: We obtain the ionic column densities from the absorption troughs in the spectrum, and use photoionization modeling to extract the physical parameters of the outflow, including the total hydrogen column density and ionization parameter. The relative population of the observed excited states of Fe II is used to model the hydrogen number density of the outflow. Results: We use the Fe II excited states to model the electron number density ($n_e$) and hydrogen number density ($n_H$) independently and obtain $n_e$ $\simeq$ $10^{3.4}$ cm$^{-3}$ and $n_H$ $\simeq$ $10^{4.8}$ cm$^{-3}$. Our analysis of the physical structure of the cloud shows that these two results are consistent with each other. This places the outflow system at a distance of $0.7_{-0.4}^{+0.9}$ kpc from the central source, with mass flow rate ($\dot{M}$) of $43^{+65}_{-26}$ $M_\odot$ yr$^{-1}$ and kinetic luminosity ($\dot{E_k}$) of $0.7^{+1.1}_{-0.4}$ $\times$ $10^{43}$ erg s$^{-1}$., Comment: 15 pages and 10 figures more...

We report on a search for millicharged particles (mCPs) produced in cosmic ray proton atmospheric interactions using data collected during the first science run of the LUX-ZEPLIN experiment. The mCPs produced by two processes -- meson decay and proton bremsstrahlung -- are considered in this study. This search utilized a novel signature unique to liquid xenon (LXe) time projection chambers (TPCs), allowing sensitivity to mCPs with masses ranging from 10 to 1000 MeV/c$^2$ and fractional charges between 0.001 and 0.02 of the electron charge e. With an exposure of 60 live days and a 5.5 tonne fiducial mass, we observed no significant excess over background. This represents the first experimental search for atmospheric mCPs and the first search for mCPs using an underground LXe experiment. more...

Large language models (LLMs) are increasingly deployed via public-facing interfaces to interact with millions of users, each with diverse preferences. Despite this, preference tuning of LLMs predominantly relies on reward models trained using binary judgments where annotators select the preferred choice out of pairs of model outputs. In this work, we argue that this reliance on binary choices does not capture the broader, aggregate preferences of the target user in real-world tasks. We propose a taxonomy that identifies two dimensions of subjectivity where different users disagree on the preferred output-namely, the Plurality of Responses to Prompts, where prompts allow for multiple correct answers, and the Indistinguishability of Responses, where candidate outputs are paraphrases of each other. We show that reward models correlate weakly with user preferences in these cases. As a first step to address this issue, we introduce a simple yet effective method that augments existing binary preference datasets with synthetic preference judgments to estimate potential user disagreement. Incorporating these via a margin term as a form of regularization during model training yields predictions that better align with the aggregate user preferences. more...

Cluster analyses of high-dimensional data are often hampered by the presence of large numbers of variables that do not provide relevant information, as well as the perennial issue of choosing an appropriate number of clusters. These challenges are frequently encountered when analysing `omics datasets, such as in molecular precision medicine, where a key goal is to identify disease subtypes and the biomarkers that define them. Here we introduce an annealed variational Bayes algorithm for fitting high-dimensional mixture models while performing variable selection. Our algorithm is scalable and computationally efficient, and we provide an open source Python implementation, VBVarSel. In a range of simulated and real biomedical examples, we show that VBVarSel outperforms the current state of the art, and demonstrate its use for cancer subtyping and biomarker discovery. more...

With the advancement of large language models (LLMs), the biomedical domain has seen significant progress and improvement in multiple tasks such as biomedical question answering, lay language summarization of the biomedical literature, clinical note summarization, etc. However, hallucinations or confabulations remain one of the key challenges when using LLMs in the biomedical and other domains. Inaccuracies may be particularly harmful in high-risk situations, such as medical question answering, making clinical decisions, or appraising biomedical research. Studies on the evaluation of the LLMs abilities to ground generated statements in verifiable sources have shown that models perform significantly worse on lay-user-generated questions, and often fail to reference relevant sources. This can be problematic when those seeking information want evidence from studies to back up the claims from LLMs. Unsupported statements are a major barrier to using LLMs in any applications that may affect health. Methods for grounding generated statements in reliable sources along with practical evaluation approaches are needed to overcome this barrier. Towards this, in our pilot task organized at TREC 2024, we introduced the task of reference attribution as a means to mitigate the generation of false statements by LLMs answering biomedical questions. more...

Legacy software systems, written in outdated languages like MUMPS and mainframe assembly, pose challenges in efficiency, maintenance, staffing, and security. While LLMs offer promise for modernizing these systems, their ability to understand legacy languages is largely unknown. This paper investigates the utilization of LLMs to generate documentation for legacy code using two datasets: an electronic health records (EHR) system in MUMPS and open-source applications in IBM mainframe Assembly Language Code (ALC). We propose a prompting strategy for generating line-wise code comments and a rubric to evaluate their completeness, readability, usefulness, and hallucination. Our study assesses the correlation between human evaluations and automated metrics, such as code complexity and reference-based metrics. We find that LLM-generated comments for MUMPS and ALC are generally hallucination-free, complete, readable, and useful compared to ground-truth comments, though ALC poses challenges. However, no automated metrics strongly correlate with comment quality to predict or measure LLM performance. Our findings highlight the limitations of current automated measures and the need for better evaluation metrics for LLM-generated documentation in legacy systems., Comment: Abbreviated version submitted to LLM4Code 2025 (a workshop co-located with ICSE 2025), 13 pages, 3 figures more...

The capabilities and limitations of Large Language Models have been sketched out in great detail in recent years, providing an intriguing yet conflicting picture. On the one hand, LLMs demonstrate a general ability to solve problems. On the other hand, they show surprising reasoning gaps when compared to humans, casting doubt on the robustness of their generalisation strategies. The sheer volume of data used in the design of LLMs has precluded us from applying the method traditionally used to measure generalisation: train-test set separation. To overcome this, we study what kind of generalisation strategies LLMs employ when performing reasoning tasks by investigating the pretraining data they rely on. For two models of different sizes (7B and 35B) and 2.5B of their pretraining tokens, we identify what documents influence the model outputs for three simple mathematical reasoning tasks and contrast this to the data that are influential for answering factual questions. We find that, while the models rely on mostly distinct sets of data for each factual question, a document often has a similar influence across different reasoning questions within the same task, indicating the presence of procedural knowledge. We further find that the answers to factual questions often show up in the most influential data. However, for reasoning questions the answers usually do not show up as highly influential, nor do the answers to the intermediate reasoning steps. When we characterise the top ranked documents for the reasoning questions qualitatively, we confirm that the influential documents often contain procedural knowledge, like demonstrating how to obtain a solution using formulae or code. Our findings indicate that the approach to reasoning the models use is unlike retrieval, and more like a generalisable strategy that synthesises procedural knowledge from documents doing a similar form of reasoning. more...

Objectives: Despite the recent adoption of large language models (LLMs) for biomedical information extraction, challenges in prompt engineering and algorithms persist, with no dedicated software available. To address this, we developed LLM-IE: a Python package for building complete information extraction pipelines. Our key innovation is an interactive LLM agent to support schema definition and prompt design. Materials and Methods: The LLM-IE supports named entity recognition, entity attribute extraction, and relation extraction tasks. We benchmarked on the i2b2 datasets and conducted a system evaluation. Results: The sentence-based prompting algorithm resulted in the best performance while requiring a longer inference time. System evaluation provided intuitive visualization. Discussion: LLM-IE was designed from practical NLP experience in healthcare and has been adopted in internal projects. It should hold great value to the biomedical NLP community. Conclusion: We developed a Python package, LLM-IE, that provides building blocks for robust information extraction pipeline construction. more...

An extreme yet reconfigurable nonlinear response to a single photon by a photonic system is crucial for realizing a universal two-photon gate, an elementary building block for photonic quantum computing. Yet such a response, characterized by the photon blockade effect, has only been achieved in atomic systems or solid states ones that are difficult to scale up. Here we demonstrate electrically tunable partial photon blockade in dipolar waveguide polaritons on a semiconductor chip, measured via photon-correlations. Remarkably, these "dipolar photons" display a two-orders-of-magnitude stronger nonlinearity compared to unpolarized polaritons, with an extracted dipolar blockade radius up to more than 4 $\mu$m, significantly larger than the optical wavelength, and comparable to that of atomic Rydberg polaritons. Furthermore, we show that the dipolar interaction can be electrically switched and locally configured by simply tuning the gate voltage. Finally we show that with a simple modification of the design, a full photon blockade is expected, setting a new route towards scalable, reconfigurable, chip-integrated quantum photonic circuits with strong two-photon nonlinearities., Comment: 30 pages, 11 figures more...

In this Letter we report on the experimental generation of high energy (10 GeV), ultra-short (fs-duration), ultra-high current (0.1 MA), petawatt peak power electron beams in a particle accelerator. These extreme beams enable the exploration of a new frontier of high intensity beam-light and beam-matter interactions broadly relevant across fields ranging from laboratory astrophysics to strong field quantum electrodynamics and ultra-fast quantum chemistry. We demonstrate our ability to generate and control the properties of these electron beams by means of a laser-electron beam shaping technique. This experimental demonstration opens the door to on-the-fly customization of extreme beam current profiles for desired experiments and is poised to benefit a broad swathe of cross-cutting applications of relativistic electron beams. more...

In the span of four decades, quantum computation has evolved from an intellectual curiosity to a potentially realizable technology. Today, small-scale demonstrations have become possible for quantum algorithmic primitives on hundreds of physical qubits and proof-of-principle error-correction on a single logical qubit. Nevertheless, despite significant progress and excitement, the path toward a full-stack scalable technology is largely unknown. There are significant outstanding quantum hardware, fabrication, software architecture, and algorithmic challenges that are either unresolved or overlooked. These issues could seriously undermine the arrival of utility-scale quantum computers for the foreseeable future. Here, we provide a comprehensive review of these scaling challenges. We show how the road to scaling could be paved by adopting existing semiconductor technology to build much higher-quality qubits, employing system engineering approaches, and performing distributed quantum computation within heterogeneous high-performance computing infrastructures. These opportunities for research and development could unlock certain promising applications, in particular, efficient quantum simulation/learning of quantum data generated by natural or engineered quantum systems. To estimate the true cost of such promises, we provide a detailed resource and sensitivity analysis for classically hard quantum chemistry calculations on surface-code error-corrected quantum computers given current, target, and desired hardware specifications based on superconducting qubits, accounting for a realistic distribution of errors. Furthermore, we argue that, to tackle industry-scale classical optimization and machine learning problems in a cost-effective manner, heterogeneous quantum-probabilistic computing with custom-designed accelerators should be considered as a complementary path toward scalability., Comment: 76 pages, 46 figures. General revision, added figures, added references, added appendices more...

Backgrounds: Information extraction (IE) is critical in clinical natural language processing (NLP). While large language models (LLMs) excel on generative tasks, their performance on extractive tasks remains debated. Methods: We investigated Named Entity Recognition (NER) and Relation Extraction (RE) using 1,588 clinical notes from four sources (UT Physicians, MTSamples, MIMIC-III, and i2b2). We developed an annotated corpus covering 4 clinical entities and 16 modifiers, and compared instruction-tuned LLaMA-2 and LLaMA-3 against BERT in terms of performance, generalizability, computational resources, and throughput to BERT. Results: LLaMA models outperformed BERT across datasets. With sufficient training data, LLaMA showed modest improvements (1% on NER, 1.5-3.7% on RE); improvements were larger with limited training data. On unseen i2b2 data, LLaMA-3-70B outperformed BERT by 7% (F1) on NER and 4% on RE. However, LLaMA models required more computing resources and ran up to 28 times slower. We implemented "Kiwi," a clinical IE package featuring both models, available at https://kiwi.clinicalnlp.org/. Conclusion: This study is among the first to develop and evaluate a comprehensive clinical IE system using open-source LLMs. Results indicate that LLaMA models outperform BERT for clinical NER and RE but with higher computational costs and lower throughputs. These findings highlight that choosing between LLMs and traditional deep learning methods for clinical IE applications should remain task-specific, taking into account both performance metrics and practical considerations such as available computing resources and the intended use case scenarios. more...

Recent advancements in artificial intelligence (AI) for numerical weather prediction (NWP) have significantly transformed atmospheric modeling. AI NWP models outperform traditional physics-based systems, such as the Integrated Forecast System (IFS), across several global metrics while requiring fewer computational resources. However, existing AI NWP models face limitations related to training datasets and timestep choices, often resulting in artifacts that reduce model performance. To address these challenges, we introduce the Community Research Earth Digital Intelligence Twin (CREDIT) framework, developed at NSF NCAR. CREDIT provides a flexible, scalable, and user-friendly platform for training and deploying AI-based atmospheric models on high-performance computing systems. It offers an end-to-end pipeline for data preprocessing, model training, and evaluation, democratizing access to advanced AI NWP capabilities. We demonstrate CREDIT's potential through WXFormer, a novel deterministic vision transformer designed to predict atmospheric states autoregressively, addressing common AI NWP issues like compounding error growth with techniques such as spectral normalization, padding, and multi-step training. Additionally, to illustrate CREDIT's flexibility and state-of-the-art model comparisons, we train the FUXI architecture within this framework. Our findings show that both FUXI and WXFormer, trained on six-hourly ERA5 hybrid sigma-pressure levels, generally outperform IFS HRES in 10-day forecasts, offering potential improvements in efficiency and forecast accuracy. CREDIT's modular design enables researchers to explore various models, datasets, and training configurations, fostering innovation within the scientific community. more...

Many workers in cities across the world, who have been teleworking because of the COVID-19 pandemic, are expected to be back to their commutes. As this process is believed to be gradual and telecommuting is likely to remain an option for many workers, hybrid model and flexible schedules might become the norm in the future. This variable work schedules allows employees to commute outside of traditional rush hours. Moreover, many studies showed that commuters might be skeptical of using trains, buses, and carpools and could turn to personal vehicles to get to work, which might increase congestion and crashes in the roads. This study attempts to provide information on the safest time to commute to Washington, DC area analyzing historical traffic crash data before the COVID-19 pandemic. It also aims to advance our understanding of traffic crashes and other relating factors such as weather in the Washington, DC area. We created a model to predict crashes by time of the day, using a negative binomial regression after rejecting a Poisson regression, and additionally explored the validity of a Random Forest regression. Our main consideration for an eventual application of this study is to reduce crashes in Washington DC, using this tool that provides people with better options on when to commute and when to telework, if available. The study also provides policymakers and researchers with real-world insights that decrease the number of traffic crashes to help achieve the goals of The Vision Zero Initiative adopted by the district. more...