Search

Context. The UVJ color-color diagram is a widely used diagnostic to separate star-forming and quiescent galaxies. Observational data from photometric surveys reveal a strong stellar mass trend, with higher-mass star-forming galaxies being systematically more dust-reddened. Aims. We analyze the UVJ diagram in the TNG100 cosmological simulation at cosmic noon ($z\approx2$). Specifically, we focus on the trend between UVJ colors and mass which has so far not been reproduced in any cosmological simulation. Methods. We apply the SKIRT dust radiative transfer code to the TNG100 simulation to generate rest-frame UVJ fluxes. These UVJ colors are then compared to observational data from several well-studied extragalactic fields from the CANDELS/3D-HST programs, augmented by recent JWST/NIRCam photometry. Results. Quiescent and low-mass ($M_\star\lesssim10^{10.5}\,\mathrm{M}_\odot$) galaxies at cosmic noon do not require significant levels of dust reddening, as opposed to massive ($M_\star\gtrsim10^{11}\,\mathrm{M}_\odot$) star-forming galaxies. An extensive range of possible dust models fall short of the required dust reddening in V-J color for massive star-forming galaxies, with the simulated galaxies being too blue by $\approx0.9\,\mathrm{mag}$. Conclusions. We find that only variations in the star-to-dust geometries of the simulated galaxies can yield V-J colors that are red enough to match the observations. A toy model with isolated dust screens around younger stellar populations (with ages below $\sim1\,\mathrm{Gyr}$) can reproduce the observational data, while all conventional dust radiative transfer models (where the dust distribution follows the metals in the interstellar medium) fail to achieve the required V-J colors., Comment: Main text 17 pages, 12 figures. Resubmitted to A&A after first referee report. Comments warmly welcome!

Rapid advancements in large language models have unlocked remarkable capabilities when it comes to processing and summarizing unstructured text data. This has implications for the analysis of rich, open-ended datasets, such as survey responses, where LLMs hold the promise of efficiently distilling key themes and sentiments. However, as organizations increasingly turn to these powerful AI systems to make sense of textual feedback, a critical question arises, can we trust LLMs to accurately represent the perspectives contained within these text based datasets? While LLMs excel at generating human-like summaries, there is a risk that their outputs may inadvertently diverge from the true substance of the original responses. Discrepancies between the LLM-generated outputs and the actual themes present in the data could lead to flawed decision-making, with far-reaching consequences for organizations. This research investigates the effectiveness of LLM-as-judge models to evaluate the thematic alignment of summaries generated by other LLMs. We utilized an Anthropic Claude model to generate thematic summaries from open-ended survey responses, with Amazon's Titan Express, Nova Pro, and Meta's Llama serving as judges. This LLM-as-judge approach was compared to human evaluations using Cohen's kappa, Spearman's rho, and Krippendorff's alpha, validating a scalable alternative to traditional human centric evaluation methods. Our findings reveal that while LLM-as-judge offer a scalable solution comparable to human raters, humans may still excel at detecting subtle, context-specific nuances. Our research contributes to the growing body of knowledge on AI assisted text analysis. Further, we provide recommendations for future research, emphasizing the need for careful consideration when generalizing LLM-as-judge models across various contexts and use cases., Comment: 11 pages, 1 appendix

This paper explores ideas and provides a potential roadmap for the development and evaluation of physics-specific large-scale AI models, which we call Large Physics Models (LPMs). These models, based on foundation models such as Large Language Models (LLMs) - trained on broad data - are tailored to address the demands of physics research. LPMs can function independently or as part of an integrated framework. This framework can incorporate specialized tools, including symbolic reasoning modules for mathematical manipulations, frameworks to analyse specific experimental and simulated data, and mechanisms for synthesizing theories and scientific literature. We begin by examining whether the physics community should actively develop and refine dedicated models, rather than relying solely on commercial LLMs. We then outline how LPMs can be realized through interdisciplinary collaboration among experts in physics, computer science, and philosophy of science. To integrate these models effectively, we identify three key pillars: Development, Evaluation, and Philosophical Reflection. Development focuses on constructing models capable of processing physics texts, mathematical formulations, and diverse physical data. Evaluation assesses accuracy and reliability by testing and benchmarking. Finally, Philosophical Reflection encompasses the analysis of broader implications of LLMs in physics, including their potential to generate new scientific understanding and what novel collaboration dynamics might arise in research. Inspired by the organizational structure of experimental collaborations in particle physics, we propose a similarly interdisciplinary and collaborative approach to building and refining Large Physics Models. This roadmap provides specific objectives, defines pathways to achieve them, and identifies challenges that must be addressed to realise physics-specific large scale AI models.

Signal Temporal Logic (STL) offers a concise yet expressive framework for specifying and reasoning about spatio-temporal behaviors of robotic systems. Attractively, STL admits the notion of robustness, the degree to which an input signal satisfies or violates an STL specification, thus providing a nuanced evaluation of system performance. Notably, the differentiability of STL robustness enables direct integration to robotics workflows that rely on gradient-based optimization, such as trajectory optimization and deep learning. However, existing approaches to evaluating and differentiating STL robustness rely on recurrent computations, which become inefficient with longer sequences, limiting their use in time-sensitive applications. In this paper, we present STLCG++, a masking-based approach that parallelizes STL robustness evaluation and backpropagation across timesteps, achieving more than 1000x faster computation time than the recurrent approach. We also introduce a smoothing technique for differentiability through time interval bounds, expanding STL's applicability in gradient-based optimization tasks over spatial and temporal variables. Finally, we demonstrate STLCG++'s benefits through three robotics use cases and provide open-source Python libraries in JAX and PyTorch for seamless integration into modern robotics workflows., Comment: To be submitted to robotics journal for review

Security-Enhanced Linux (SELinux) is a robust security mechanism that enforces mandatory access controls (MAC), but its policy language's complexity creates challenges for policy analysis and management. This research investigates the automation of SELinux policy analysis using graph-based techniques combined with machine learning approaches to detect policy anomalies. The study addresses two key questions: Can SELinux policy analysis be automated through graph analysis, and how do different anomaly detection models compare in analyzing SELinux policies? We will be comparing different machine learning models by evaluating their effectiveness in detecting policy violations and anomalies. Our approach utilizes Neo4j for graph representation of policies, with Node2vec transforming these graph structures into meaningful vector embeddings that can be processed by our machine learning models. In our results, the MLP Neural Network consistently demonstrated superior performance across different dataset sizes, achieving 95% accuracy with balanced precision and recall metrics, while both Random Forest and SVM models showed competitive but slightly lower performance in detecting policy violations. This combination of graph-based modeling and machine learning provides a more sophisticated and automated approach to understanding and analyzing complex SELinux policies compared to traditional manual analysis methods.

Fantasy sports, particularly fantasy cricket, have garnered immense popularity in India in recent years, offering enthusiasts the opportunity to engage in strategic team-building and compete based on the real-world performance of professional athletes. In this paper, we address the challenge of optimizing fantasy cricket team selection using reinforcement learning (RL) techniques. By framing the team creation process as a sequential decision-making problem, we aim to develop a model that can adaptively select players to maximize the team's potential performance. Our approach leverages historical player data to train RL algorithms, which then predict future performance and optimize team composition. This not only represents a huge business opportunity by enabling more accurate predictions of high-performing teams but also enhances the overall user experience. Through empirical evaluation and comparison with traditional fantasy team drafting methods, we demonstrate the effectiveness of RL in constructing competitive fantasy teams. Our results show that RL-based strategies provide valuable insights into player selection in fantasy sports., Comment: 8 Pages including references, Accepted to CODS-COMAD 2024 conference

In multi-agent environments, agents often struggle to learn optimal policies due to sparse or delayed global rewards, particularly in long-horizon tasks where it is challenging to evaluate actions at intermediate time steps. We introduce Temporal-Agent Reward Redistribution (TAR$^2$), a novel approach designed to address the agent-temporal credit assignment problem by redistributing sparse rewards both temporally and across agents. TAR$^2$ decomposes sparse global rewards into time-step-specific rewards and calculates agent-specific contributions to these rewards. We theoretically prove that TAR$^2$ is equivalent to potential-based reward shaping, ensuring that the optimal policy remains unchanged. Empirical results demonstrate that TAR$^2$ stabilizes and accelerates the learning process. Additionally, we show that when TAR$^2$ is integrated with single-agent reinforcement learning algorithms, it performs as well as or better than traditional multi-agent reinforcement learning methods., Comment: 12 pages, 1 figure

In the evolving landscape of machine learning (ML), Federated Learning (FL) presents a paradigm shift towards decentralized model training while preserving user data privacy. This paper introduces the concept of ``privacy drift", an innovative framework that parallels the well-known phenomenon of concept drift. While concept drift addresses the variability in model accuracy over time due to changes in the data, privacy drift encapsulates the variation in the leakage of private information as models undergo incremental training. By defining and examining privacy drift, this study aims to unveil the nuanced relationship between the evolution of model performance and the integrity of data privacy. Through rigorous experimentation, we investigate the dynamics of privacy drift in FL systems, focusing on how model updates and data distribution shifts influence the susceptibility of models to privacy attacks, such as membership inference attacks (MIA). Our results highlight a complex interplay between model accuracy and privacy safeguards, revealing that enhancements in model performance can lead to increased privacy risks. We provide empirical evidence from experiments on customized datasets derived from CIFAR-100 (Canadian Institute for Advanced Research, 100 classes), showcasing the impact of data and concept drift on privacy. This work lays the groundwork for future research on privacy-aware machine learning, aiming to achieve a delicate balance between model accuracy and data privacy in decentralized environments., Comment: 6 pages, 7 figures, Accepted in IEEE ICNC 25

Temporal logic specifications play an important role in a wide range of software analysis tasks, such as model checking, automated synthesis, program comprehension, and runtime monitoring. Given a set of positive and negative examples, specified as traces, LTL learning is the problem of synthesizing a specification, in linear temporal logic (LTL), that evaluates to true over the positive traces and false over the negative ones. In this paper, we propose a new type of LTL learning problem called constrained LTL learning, where the user, in addition to positive and negative examples, is given an option to specify one or more constraints over the properties of the LTL formula to be learned. We demonstrate that the ability to specify these additional constraints significantly increases the range of applications for LTL learning, and also allows efficient generation of LTL formulas that satisfy certain desirable properties (such as minimality). We propose an approach for solving the constrained LTL learning problem through an encoding in first-order relational logic and reduction to an instance of the maximal satisfiability (MaxSAT) problem. An experimental evaluation demonstrates that ATLAS, an implementation of our proposed approach, is able to solve new types of learning problems while performing better than or competitively with the state-of-the-art tools in LTL learning., Comment: 14 pages, ICSE 2025

To accurately and confidently answer the question 'could an AI model or system increase biorisk', it is necessary to have both a sound theoretical threat model for how AI models or systems could increase biorisk and a robust method for testing that threat model. This paper provides an analysis of existing available research surrounding two AI and biorisk threat models: 1) access to information and planning via large language models (LLMs), and 2) the use of AI-enabled biological tools (BTs) in synthesizing novel biological artifacts. We find that existing studies around AI-related biorisk are nascent, often speculative in nature, or limited in terms of their methodological maturity and transparency. The available literature suggests that current LLMs and BTs do not pose an immediate risk, and more work is needed to develop rigorous approaches to understanding how future models could increase biorisks. We end with recommendations about how empirical work can be expanded to more precisely target biorisk and ensure rigor and validity of findings., Comment: Updated to correct author affiliations and clarify findings of evaluations of the o1 model

Recent research has generated hope that inference scaling could allow weaker language models to match or exceed the accuracy of stronger models, such as by repeatedly sampling solutions to a coding problem until it passes unit tests. The central thesis of this paper is that there is no free lunch for inference scaling: indefinite accuracy improvement through resampling can only be realized if the "verifier" (in this case, a set of unit tests) is perfect. When the verifier is imperfect, as it almost always is in domains such as reasoning or coding (for example, unit tests have imperfect coverage), there is a nonzero probability of false positives: incorrect solutions that pass the verifier. Resampling cannot decrease this probability, so it imposes an upper bound to the accuracy of resampling-based inference scaling even with an infinite compute budget. We find that there is a very strong correlation between the model's single-sample accuracy (i.e. accuracy without unit tests) and its false positive rate on coding benchmarks HumanEval and MBPP, whose unit tests have limited coverage. Therefore, no amount of inference scaling of weaker models can enable them to match the single-sample accuracy of a sufficiently strong model (Fig. 1a). When we consider that false positives have a negative utility compared to abstaining from producing a solution, it bends the inference scaling curve further downward. Empirically, we find that the optimal number of samples can be less than 10 under realistic assumptions (Fig. 1b). Finally, we show that beyond accuracy, false positives may have other undesirable qualities, such as poor adherence to coding style conventions.

Scientists and engineers use simulators to model empirically observed phenomena. However, tuning the parameters of a simulator to ensure its outputs match observed data presents a significant challenge. Simulation-based inference (SBI) addresses this by enabling Bayesian inference for simulators, identifying parameters that match observed data and align with prior knowledge. Unlike traditional Bayesian inference, SBI only needs access to simulations from the model and does not require evaluations of the likelihood-function. In addition, SBI algorithms do not require gradients through the simulator, allow for massive parallelization of simulations, and can perform inference for different observations without further simulations or training, thereby amortizing inference. Over the past years, we have developed, maintained, and extended $\texttt{sbi}$, a PyTorch-based package that implements Bayesian SBI algorithms based on neural networks. The $\texttt{sbi}$ toolkit implements a wide range of inference methods, neural network architectures, sampling methods, and diagnostic tools. In addition, it provides well-tested default settings but also offers flexibility to fully customize every step of the simulation-based inference workflow. Taken together, the $\texttt{sbi}$ toolkit enables scientists and engineers to apply state-of-the-art SBI methods to black-box simulators, opening up new possibilities for aligning simulations with empirically observed data.

We propose PPLqa, an easy to compute, language independent, information-theoretic metric to measure the quality of responses of generative Large Language Models (LLMs) in an unsupervised way, without requiring ground truth annotations or human supervision. The method and metric enables users to rank generative language models for quality of responses, so as to make a selection of the best model for a given task. Our single metric assesses LLMs with an approach that subsumes, but is not explicitly based on, coherence and fluency (quality of writing) and relevance and consistency (appropriateness of response) to the query. PPLqa performs as well as other related metrics, and works better with long-form Q\&A. Thus, PPLqa enables bypassing the lengthy annotation process required for ground truth evaluations, and it also correlates well with human and LLM rankings.

In this Technical Design Report (TDR) we describe the SABRE South detector to be built at the Stawell Underground Physics Laboratory (SUPL). The SABRE South detector is designed to test the long-standing DAMA/LIBRA signal of an annually modulating rate consistent with dark matter by using the same target material. SABRE South uses seven ultra-high purity NaI(Tl) crystals (with a total target mass of either 35 kg or 50 kg), hermetically sealed in copper enclosures that are suspended within a liquid scintillator active veto. High quantum efficiency and low background Hamamatsu R11065 photomultiplier tubes are directly coupled to both ends of the crystal, and enclosed with the crystal in an oxygen free high thermal conductivity copper enclosure. The active veto system consists of 11.6 kL of linear alkylbenzene (LAB) doped with a mixture of fluorophores and contained in a steel vessel, which is instrumented with at least 18 Hamamatsu R5912 photomultipliers. The active veto tags key radiogenic backgrounds intrinsic to the crystals, such as ${^{40}}$K, and is expected to suppress the total background by 27% in the 1-6 keV region of interest. In addition to the liquid scintillator veto, a muon veto is positioned above the detector shielding. This muon veto consists of eight EJ-200 scintillator modules, with Hamamatsu R13089 photomultipliers coupled to both ends. With an expected total background of 0.72 cpd/kg/keV, SABRE South can test the DAMA/LIBRA signal with 5$\sigma$ discovery or 3$\sigma$ exclusion after two years of data taking.

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

This research work introduces a novel approach to the classification of Alzheimer's disease by using the advanced deep learning techniques combined with secure data processing methods. This research work primary uses transfer learning models such as ResNet, ImageNet, and VNet to extract high-level features from medical image data. Thereafter, these pre-trained models were fine-tuned for Alzheimer's related subtle patterns such that the model is capable of robust feature extraction over varying data sources. Further, the federated learning approaches were incorporated to tackle a few other challenges related to classification, aimed to provide better prediction performance and protect data privacy. The proposed model was built using federated learning without sharing sensitive patient data. This way, the decentralized model benefits from the large and diversified dataset that it is trained upon while ensuring confidentiality. The cipher-based encryption mechanism is added that allows us to secure the transportation of data and further ensure the privacy and integrity of patient information throughout training and classification. The results of the experiments not only help to improve the accuracy of the classification of Alzheimer's but at the same time provides a framework for secure and collaborative analysis of health care data., Comment: 6 pages, 4 figures

The study of $\bar{B} \to D^* \tau {\bar{\nu}}_\tau$ angular distribution can be used to obtain information about new physics (or beyond the Standard Model) couplings, which are motivated by various $B$ anomalies. However, the inability to measure precisely the three-momentum of the $\tau$ lepton hinders such measurements, as the tau decay contains one or more undetected neutrinos. Here, we present a measurable angular distribution of $\bar{B} \to D^* \tau {\bar{\nu}}_\tau$ by considering the additional decay $\tau \to \ell \nu_\tau \bar{\nu}_\ell$, where $\ell \in \{ e , \mu \}$. The full process used is $\bar{B} \to D^* (\to D \pi) \tau (\to \ell \nu_\tau \bar{\nu}_\ell) \bar{\nu}_\tau$, in which only the $\ell$ and $D^*$ are reconstructed. A fit to the experimental angular distribution of this process can be used to extract information on new physics parameters. To demonstrate the feasibility of this approach, we generate simulated data for this process and perform a sensitivity study to obtain the expected statistical errors on new physics parameters from experiments in the near future. We obtain a sensitivity of the order of 5% for the right-handed current and around 6% for the tensor current. In addition, we use the recent lattice QCD data on $B \to D^*$ form factors and obtain correlations between form factors and new physics parameters., Comment: 38 pages, 8 figures; new version: added references, corrected typos

Aggregating large-scale radiotherapy planning and delivery data is crucial for advancing radiation oncology research and improving clinical practice, yet challenges persist due to the diversity of treatment planning systems (TPS), record and verify (R&V) systems, and complex data formats lacking standardized retrieval methods. We developed a robust software framework that automates the collection and integration of multi-institutional radiotherapy data from diverse TPS and R&V systems. By utilizing the unidirectional references of DICOM objects, our framework reconstructs complete patient datasets starting from Radiotherapy Treatment Records (RTRECORDs), managing tasks such as data queries, transfers, verification, and logging. It effectively maps DICOM linkages between RTRECORDs, RTPLANs, RTDOSEs, RTSTRUCTs, planning images, registrations, and associated diagnostic images, incorporating custom modules for data conversion and comprehensive error handling. Implemented across multiple institutions using various systems$-$ including ARIA, Eclipse, MOSAIQ, RayStation, MIM, Pinnacle$-$ the framework successfully collected data from two clinics over an 11-year period, aggregating data from 6,022 patients and 13,871 treatment plans with a success rate of 99.76% and an average processing time of approximately 18 minutes per patient. Ongoing efforts are extending data collection to clinics lacking DICOM Query/Retrieve capabilities, demonstrating the framework's adaptability to various clinical environments. This efficient automation of comprehensive data collection overcomes significant technical barriers, facilitating the creation of large-scale datasets that can accelerate advancements in radiation oncology., Comment: 11 pages, 5 figures

Sterile neutrinos are well-motivated beyond the Standard Model (BSM) particles. The Standard Model Effective Field Theory (SMEFT) augmented with these new fields is known as the $\nu$SMEFT. We present the first code for solving the renormalization group equations (RGEs) of the $\nu$SMEFT in an automated way. For this purpose, we have implemented the $\nu$SMEFT as a new effective field theory (EFT) in the Wilson coefficient exchange format WCxf. Furthermore, we included anomalous dimensions depending on the gauge couplings and Yukawas in the python package wilson. This novel version of wilson allows a consistent inclusion of $\nu$SMEFT renormalization group (RG) running effects above the electroweak (EW) scale in phenomenological studies involving sterile neutrinos. Moreover, this new release allows us to study EW, strong, and Yukawa running effects separately within the SMEFT., Comment: 14 pages

This paper explores the transformative impact of Generative AI (GenAI) on the business landscape, examining its role in reshaping traditional business models, intensifying market competition, and fostering innovation. By applying the principles of Neo-Schumpeterian economics, the research analyses how GenAI is driving a new wave of "creative destruction," leading to the emergence of novel business paradigms and value propositions. The findings reveal that GenAI enhances operational efficiency, facilitates product and service innovation, and creates new revenue streams, positioning it as a powerful catalyst for substantial shifts in business structures and strategies. However, the deployment of GenAI also presents significant challenges, including ethical concerns, regulatory demands, and the risk of job displacement. By addressing the multifarious nature of GenAI, this paper provides valuable insights for business leaders, policymakers, and researchers, guiding them towards a balanced and responsible integration of this transformative technology. Ultimately, GenAI is not merely a technological advancement but a driver of profound change, heralding a future where creativity, efficiency, and growth are redefined., Comment: 30 pages, 12 figures, original submission

Delirium is an acute confusional state that has been shown to affect up to 31% of patients in the intensive care unit (ICU). Early detection of this condition could lead to more timely interventions and improved health outcomes. While artificial intelligence (AI) models have shown great potential for ICU delirium prediction using structured electronic health records (EHR), most of them have not explored the use of state-of-the-art AI models, have been limited to single hospitals, or have been developed and validated on small cohorts. The use of large language models (LLM), models with hundreds of millions to billions of parameters, with structured EHR data could potentially lead to improved predictive performance. In this study, we propose DeLLiriuM, a novel LLM-based delirium prediction model using EHR data available in the first 24 hours of ICU admission to predict the probability of a patient developing delirium during the rest of their ICU admission. We develop and validate DeLLiriuM on ICU admissions from 104,303 patients pertaining to 195 hospitals across three large databases: the eICU Collaborative Research Database, the Medical Information Mart for Intensive Care (MIMIC)-IV, and the University of Florida Health's Integrated Data Repository. The performance measured by the area under the receiver operating characteristic curve (AUROC) showed that DeLLiriuM outperformed all baselines in two external validation sets, with 0.77 (95% confidence interval 0.76-0.78) and 0.84 (95% confidence interval 0.83-0.85) across 77,543 patients spanning 194 hospitals. To the best of our knowledge, DeLLiriuM is the first LLM-based delirium prediction tool for the ICU based on structured EHR data, outperforming deep learning baselines which employ structured features and can provide helpful information to clinicians for timely interventions.

State Space Models (SSMs) like Mamba2 are a promising alternative to Transformers, with faster theoretical training and inference times -- especially for long context lengths. Recent work on Matryoshka Representation Learning -- and its application to Transformer backbones in works like MatFormer -- showed how to introduce nested granularities of smaller submodels in one universal elastic model. In this work, we present MatMamba: a state space model which combines Matryoshka-style learning with Mamba2, by modifying the block to contain nested dimensions to enable joint training and adaptive inference. MatMamba allows for efficient and adaptive deployment across various model sizes. We train a single large MatMamba model and are able to get a number of smaller nested models for free -- while maintaining or improving upon the performance of a baseline smaller model trained from scratch. We train language and image models at a variety of parameter sizes from 35M to 1.4B. Our results on ImageNet and FineWeb show that MatMamba models scale comparably to Transformers, while having more efficient inference characteristics. This makes MatMamba a practically viable option for deploying large-scale models in an elastic way based on the available inference compute. Code and models are open sourced at \url{https://github.com/ScaledFoundations/MatMamba}, Comment: 10 pages, 7 figures

We present a quantum network that distributes entangled photons between the University of Illinois Urbana-Champaign and a public library in Urbana. The network allows members of the public to perform measurements on the photons. We describe its design and implementation and outreach based on the network. Over 400 instances of public interaction have been logged with the system since it was launched in November 2023., Comment: 7 Pages, 8 Figures

Current galaxy formation simulations often approximate star-formation, necessitating models of star-forming regions to produce observables. In the first paper of the series, we introduced TODDLERS, a time-resolved model of UV-mm emission from star-forming regions implemented in the radiative transfer code SKIRT. This work uses SKIRT-TODDLERS to produce synthetic observations, demonstrating its potential through observables related to star-formation and comparing results with existing models in SKIRT. We calculate broadband and line emission maps for 30 Milky Way-like galaxies from the Auriga simulation at z=0. Analyzing FUV and IR data, we calculate kpc-resolved IR correction factors (k_IR), quantifying the ratio of FUV luminosity absorbed by dust to reprocessed IR luminosity. We use IR maps to calculate kpc-scale MIR (8 micron / 24 micron) and FIR (70 micron / 500 micron) colors. H alpha and H beta line maps are used to study the Balmer decrement and dust correction. We also verify the fidelity of our model FIR fine structure lines as SFR indicators. We find that the Auriga integrated UV-mm SEDs show higher FUV/NUV attenuation and lower 24 micron emission when using TODDLERS instead of the existing models in SKIRT, alleviating tensions with observations. The light-weighted mean k_IR increases with aperture and inclination, while its correlation with kpc-resolved sSFR is weaker than literature values. Kpc-scale MIR-FIR colors agree excellently with local observations, with anti-correlation degree varying by galaxy morphology. The Balmer decrement effectively corrects for dust, with attenuation law varying with dust amount. H alpha emission attenuation levels are comparable to high-density regions in state-of-the-art simulations. FIR fine-structure line emission-based luminosity-SFR relations align with global observations, [CII] showing best agreement., Comment: 22 pages, 20 figures. Accepted for publication in Astronomy and Astrophysics on October 21, 2024

Intra-cardiac echocardiography (ICE) is a crucial imaging modality used in electrophysiology (EP) and structural heart disease (SHD) interventions, providing realtime, high-resolution views from within the heart. Despite its advantages, effective manipulation of the ICE catheter requires significant expertise, which can lead to inconsistent outcomes, especially among less experienced operators. To address this challenge, we propose an AIdriven view guidance system that operates in a continuous closed-loop with human-in-the-loop feedback, designed to assist users in navigating ICE imaging without requiring specialized knowledge. Specifically, our method models the relative position and orientation vectors between arbitrary views and clinically defined ICE views in a spatial coordinate system. It guides users on how to manipulate the ICE catheter to transition from the current view to the desired view over time. By operating in a closedloop configuration, the system continuously predicts and updates the necessary catheter manipulations, ensuring seamless integration into existing clinical workflows. The effectiveness of the proposed system is demonstrated through a simulation-based performance evaluation using real clinical data, achieving an 89% success rate with 6,532 test cases. Additionally, a semi-simulation experiment with human-in-the-loop testing validated the feasibility of continuous yet discrete guidance. These results underscore the potential of the proposed method to enhance the accuracy and efficiency of ICE imaging procedures.

Topic modeling is a widely used technique for uncovering thematic structures from large text corpora. However, most topic modeling approaches e.g. Latent Dirichlet Allocation (LDA) struggle to capture nuanced semantics and contextual understanding required to accurately model complex narratives. Recent advancements in this area include methods like BERTopic, which have demonstrated significantly improved topic coherence and thus established a new standard for benchmarking. In this paper, we present a novel approach, the Qualitative Insights Tool (QualIT) that integrates large language models (LLMs) with existing clustering-based topic modeling approaches. Our method leverages the deep contextual understanding and powerful language generation capabilities of LLMs to enrich the topic modeling process using clustering. We evaluate our approach on a large corpus of news articles and demonstrate substantial improvements in topic coherence and topic diversity compared to baseline topic modeling techniques. On the 20 ground-truth topics, our method shows 70% topic coherence (vs 65% & 57% benchmarks) and 95.5% topic diversity (vs 85% & 72% benchmarks). Our findings suggest that the integration of LLMs can unlock new opportunities for topic modeling of dynamic and complex text data, as is common in talent management research contexts., Comment: 6 pages, 4 tables, 1 figure

AI agents have the potential to aid users on a variety of consequential tasks, including conducting scientific research. To spur the development of useful agents, we need benchmarks that are challenging, but more crucially, directly correspond to real-world tasks of interest. This paper introduces such a benchmark, designed to measure the accuracy of AI agents in tackling a crucial yet surprisingly challenging aspect of scientific research: computational reproducibility. This task, fundamental to the scientific process, involves reproducing the results of a study using the provided code and data. We introduce CORE-Bench (Computational Reproducibility Agent Benchmark), a benchmark consisting of 270 tasks based on 90 scientific papers across three disciplines (computer science, social science, and medicine). Tasks in CORE-Bench consist of three difficulty levels and include both language-only and vision-language tasks. We provide an evaluation system to measure the accuracy of agents in a fast and parallelizable way, saving days of evaluation time for each run compared to a sequential implementation. We evaluated two baseline agents: the general-purpose AutoGPT and a task-specific agent called CORE-Agent. We tested both variants using two underlying language models: GPT-4o and GPT-4o-mini. The best agent achieved an accuracy of 21% on the hardest task, showing the vast scope for improvement in automating routine scientific tasks. Having agents that can reproduce existing work is a necessary step towards building agents that can conduct novel research and could verify and improve the performance of other research agents. We hope that CORE-Bench can improve the state of reproducibility and spur the development of future research agents., Comment: Benchmark harness and code available at http://github.com/siegelz/core-bench

Hunger crises are critical global issues affecting millions, particularly in low-income and developing countries. This research investigates how machine learning can be utilized to predict and inform decisions regarding famine and hunger crises. By leveraging a diverse set of variables (natural, economic, and conflict-related), three machine learning models (Linear Regression, XGBoost, and RandomForestRegressor) were employed to predict food consumption scores, a key indicator of household nutrition. The RandomForestRegressor emerged as the most accurate model, with an average prediction error of 10.6%, though accuracy varied significantly across countries, ranging from 2% to over 30%. Notably, economic indicators were consistently the most significant predictors of average household nutrition, while no single feature dominated across all regions, underscoring the necessity for comprehensive data collection and tailored, country-specific models. These findings highlight the potential of machine learning, particularly Random Forests, to enhance famine prediction, suggesting that continued research and improved data gathering are essential for more effective global hunger forecasting., Comment: 17 pages, 7 figures, 2 tables

With the advent of large foundation model based planning, there is a dire need to ensure their output aligns with the stakeholder's intent. When these models are deployed in the real world, the need for alignment is magnified due to the potential cost to life and infrastructure due to unexpected faliures. Temporal Logic specifications have long provided a way to constrain system behaviors and are a natural fit for these use cases. In this work, we propose a novel approach to factor in signal temporal logic specifications while using autoregressive transformer models for trajectory planning. We also provide a trajectory dataset for pretraining and evaluating foundation models. Our proposed technique acheives 74.3 % higher specification satisfaction over the baselines., Comment: Robotics Science and Systems: Towards Safe Autonomy

Qualitative data collection and analysis approaches, such as those employing interviews and focus groups, provide rich insights into customer attitudes, sentiment, and behavior. However, manually analyzing qualitative data requires extensive time and effort to identify relevant topics and thematic insights. This study proposes a novel approach to address this challenge by leveraging Retrieval Augmented Generation (RAG) based Large Language Models (LLMs) for analyzing interview transcripts. The novelty of this work lies in strategizing the research inquiry as one that is augmented by an LLM that serves as a novice research assistant. This research explores the mental model of LLMs to serve as novice qualitative research assistants for researchers in the talent management space. A RAG-based LLM approach is extended to enable topic modeling of semi-structured interview data, showcasing the versatility of these models beyond their traditional use in information retrieval and search. Our findings demonstrate that the LLM-augmented RAG approach can successfully extract topics of interest, with significant coverage compared to manually generated topics from the same dataset. This establishes the viability of employing LLMs as novice qualitative research assistants. Additionally, the study recommends that researchers leveraging such models lean heavily on quality criteria used in traditional qualitative research to ensure rigor and trustworthiness of their approach. Finally, the paper presents key recommendations for industry practitioners seeking to reconcile the use of LLMs with established qualitative research paradigms, providing a roadmap for the effective integration of these powerful, albeit novice, AI tools in the analysis of qualitative datasets within talent, Comment: Accepted to KDD '24 workshop on Talent Management and Computing (TMC 2024). 9 pages

The persistence of the DAMA/LIBRA (DAMA) modulation over the past two decades has been a source of great contention within the dark matter community. The DAMA collaboration reports a persistent, modulating event rate within their setup of NaI(Tl) scintillating crystals at the INFN Laboratori Nazionali del Gran Sasso (LNGS) underground laboratory. A recent work alluded that this signal could have arisen due to an analysis artefact, caused by DAMA not accounting for time variation of decaying background radioisotopes in their analysis procedure. In this work, we examine in detail this 'induced modulation' effect, arguing that a number of aspects of the DAMA signal are incompatible with an induced modulation arising from decays of background isotopes over the lifetime of the experiment. Using a toy model of the DAMA/LIBRA experiment, we explore the induced modulation effect under different variations of the activities of the relevant isotopes - namely, $^3$H and $^{210}$Pb - highlighting the various inconsistencies between the resultant toy datasets and the DAMA signal. We stress the importance of the SABRE experiment, whose goal is to unambiguously test for the presence of such a modulating signal in an experiment using the same target material and comparable levels of background.

Multi-agent proximal policy optimization (MAPPO) has recently demonstrated state-of-the-art performance on challenging multi-agent reinforcement learning tasks. However, MAPPO still struggles with the credit assignment problem, wherein the sheer difficulty in ascribing credit to individual agents' actions scales poorly with team size. In this paper, we propose a multi-agent reinforcement learning algorithm that adapts recent developments in credit assignment to improve upon MAPPO. Our approach leverages partial reward decoupling (PRD), which uses a learned attention mechanism to estimate which of a particular agent's teammates are relevant to its learning updates. We use this estimate to dynamically decompose large groups of agents into smaller, more manageable subgroups. We empirically demonstrate that our approach, PRD-MAPPO, decouples agents from teammates that do not influence their expected future reward, thereby streamlining credit assignment. We additionally show that PRD-MAPPO yields significantly higher data efficiency and asymptotic performance compared to both MAPPO and other state-of-the-art methods across several multi-agent tasks, including StarCraft II. Finally, we propose a version of PRD-MAPPO that is applicable to \textit{shared} reward settings, where PRD was previously not applicable, and empirically show that this also leads to performance improvements over MAPPO., Comment: 20 pages, 5 figures, 12 tables, Reinforcement Learning Journal and Reinforcement Learning Conference 2024

Distortion identification and rectification in images and videos is vital for achieving good performance in downstream vision applications. Instead of relying on fixed trial-and-error based image processing pipelines, we propose a two-level sequential planning approach for automated image distortion classification and rectification. At the higher level it detects the class of corruptions present in the input image, if any. The lower level selects a specific algorithm to be applied, from a set of externally provided candidate algorithms. The entire two-level setup runs in the form of a single forward pass during inference and it is to be queried iteratively until the retrieval of the original image. We demonstrate improvements compared to three baselines on the object detection task on COCO image dataset with rich set of distortions. The advantage of our approach is its dynamic reconfiguration, conditioned on the input image and generalisability to unseen candidate algorithms at inference time, since it relies only on the comparison of their output of the image embeddings., Comment: 7 pages, 3 figures

Foundation models are increasingly consequential yet extremely opaque. To characterize the status quo, the Foundation Model Transparency Index was launched in October 2023 to measure the transparency of leading foundation model developers. The October 2023 Index (v1.0) assessed 10 major foundation model developers (e.g. OpenAI, Google) on 100 transparency indicators (e.g. does the developer disclose the wages it pays for data labor?). At the time, developers publicly disclosed very limited information with the average score being 37 out of 100. To understand how the status quo has changed, we conduct a follow-up study (v1.1) after 6 months: we score 14 developers against the same 100 indicators. While in v1.0 we searched for publicly available information, in v1.1 developers submit reports on the 100 transparency indicators, potentially including information that was not previously public. We find that developers now score 58 out of 100 on average, a 21 point improvement over v1.0. Much of this increase is driven by developers disclosing information during the v1.1 process: on average, developers disclosed information related to 16.6 indicators that was not previously public. We observe regions of sustained (i.e. across v1.0 and v1.1) and systemic (i.e. across most or all developers) opacity such as on copyright status, data access, data labor, and downstream impact. We publish transparency reports for each developer that consolidate information disclosures: these reports are based on the information disclosed to us via developers. Our findings demonstrate that transparency can be improved in this nascent ecosystem, the Foundation Model Transparency Index likely contributes to these improvements, and policymakers should consider interventions in areas where transparency has not improved., Comment: Authored by the Center for Research on Foundation Models (CRFM) at the Stanford Institute for Human-Centered Artificial Intelligence (HAI). Project page: https://crfm.stanford.edu/fmti

The ability of Language Models (LMs) to understand natural language makes them a powerful tool for parsing human instructions into task plans for autonomous robots. Unlike traditional planning methods that rely on domain-specific knowledge and handcrafted rules, LMs generalize from diverse data and adapt to various tasks with minimal tuning, acting as a compressed knowledge base. However, LMs in their standard form face challenges with long-horizon tasks, particularly in partially observable multi-agent settings. We propose an LM-based Long-Horizon Planner for Multi-Agent Robotics (LLaMAR), a cognitive architecture for planning that achieves state-of-the-art results in long-horizon tasks within partially observable environments. LLaMAR employs a plan-act-correct-verify framework, allowing self-correction from action execution feedback without relying on oracles or simulators. Additionally, we present MAP-THOR, a comprehensive test suite encompassing household tasks of varying complexity within the AI2-THOR environment. Experiments show that LLaMAR achieves a 30% higher success rate than other state-of-the-art LM-based multi-agent planners in MAP-THOR and Search \& Rescue tasks. Code can be found at https://github.com/nsidn98/LLaMAR, Comment: 27 pages, 4 figures, 5 tables

Knowledge Graph Embedding (KGE) transforms a discrete Knowledge Graph (KG) into a continuous vector space facilitating its use in various AI-driven applications like Semantic Search, Question Answering, or Recommenders. While KGE approaches are effective in these applications, most existing approaches assume that all information in the given KG is correct. This enables attackers to influence the output of these approaches, e.g., by perturbing the input. Consequently, the robustness of such KGE approaches has to be addressed. Recent work focused on adversarial attacks. However, non-adversarial attacks on all attack surfaces of these approaches have not been thoroughly examined. We close this gap by evaluating the impact of non-adversarial attacks on the performance of 5 state-of-the-art KGE algorithms on 5 datasets with respect to attacks on 3 attack surfaces-graph, parameter, and label perturbation. Our evaluation results suggest that label perturbation has a strong effect on the KGE performance, followed by parameter perturbation with a moderate and graph with a low effect.

AI agents are an exciting new research direction, and agent development is driven by benchmarks. Our analysis of current agent benchmarks and evaluation practices reveals several shortcomings that hinder their usefulness in real-world applications. First, there is a narrow focus on accuracy without attention to other metrics. As a result, SOTA agents are needlessly complex and costly, and the community has reached mistaken conclusions about the sources of accuracy gains. Our focus on cost in addition to accuracy motivates the new goal of jointly optimizing the two metrics. We design and implement one such optimization, showing its potential to greatly reduce cost while maintaining accuracy. Second, the benchmarking needs of model and downstream developers have been conflated, making it hard to identify which agent would be best suited for a particular application. Third, many agent benchmarks have inadequate holdout sets, and sometimes none at all. This has led to agents that are fragile because they take shortcuts and overfit to the benchmark in various ways. We prescribe a principled framework for avoiding overfitting. Finally, there is a lack of standardization in evaluation practices, leading to a pervasive lack of reproducibility. We hope that the steps we introduce for addressing these shortcomings will spur the development of agents that are useful in the real world and not just accurate on benchmarks.

Educating students about academic integrity expectations has been suggested as one of the ways to reduce malpractice in take-home programming assignments. We test this hypothesis using data collected from an artificial intelligence course with 105 participants (N=105) at a university in India. The AI course had two programming assignments. Plagiarism through collusion was quantified using the Measure of Software Similarity (MOSS) tool. Students were educated about what constitutes academic dishonesty and were required to take an honor pledge before the start of the second take-home programming assignment. The two programming assignments were novel and did not have solutions available on the internet. We expected the mean percentage of similar lines of code to be significantly less in the second programming assignment. However, our results show no significant difference in the mean percentage of similar lines of code across the two programming assignments. We also study how the Big-five personality traits affect the propensity for plagiarism in the two take-home assignments. Our results across both assignments show that the extraversion trait of the Big Five personality exhibits a positive association, and the conscientiousness trait exhibits a negative association with plagiarism tendencies. Our result suggests that the policy of educating students about academic integrity will have a limited impact as long as students perceive an opportunity for plagiarism to be present. We explain our results using the Fraud triangle model., Comment: 11 pages, 3 figures, To be published in ACM International Conference on Computing Education Research (ICER) 2024

Modern datasets in neuroscience enable unprecedented inquiries into the relationship between complex behaviors and the activity of many simultaneously recorded neurons. While latent variable models can successfully extract low-dimensional embeddings from such recordings, using them to generate realistic spiking data, especially in a behavior-dependent manner, still poses a challenge. Here, we present Latent Diffusion for Neural Spiking data (LDNS), a diffusion-based generative model with a low-dimensional latent space: LDNS employs an autoencoder with structured state-space (S4) layers to project discrete high-dimensional spiking data into continuous time-aligned latents. On these inferred latents, we train expressive (conditional) diffusion models, enabling us to sample neural activity with realistic single-neuron and population spiking statistics. We validate LDNS on synthetic data, accurately recovering latent structure, firing rates, and spiking statistics. Next, we demonstrate its flexibility by generating variable-length data that mimics human cortical activity during attempted speech. We show how to equip LDNS with an expressive observation model that accounts for single-neuron dynamics not mediated by the latent state, further increasing the realism of generated samples. Finally, conditional LDNS trained on motor cortical activity during diverse reaching behaviors can generate realistic spiking data given reach direction or unseen reach trajectories. In summary, LDNS simultaneously enables inference of low-dimensional latents and realistic conditional generation of neural spiking datasets, opening up further possibilities for simulating experimentally testable hypotheses., Comment: 38th Conference on Neural Information Processing Systems (NeurIPS 2024)