Search

Effective communication is essential in collaborative tasks, so AI-equipped robots working alongside humans need to be able to explain their behaviour in order to cooperate effectively and earn trust. We analyse and classify communications among human participants collaborating to complete a simulated emergency response task. The analysis identifies messages that relate to various kinds of interactive explanations identified in the explainable AI literature. This allows us to understand what type of explanations humans expect from their teammates in such settings, and thus where AI-equipped robots most need explanation capabilities. We find that most explanation-related messages seek clarification in the decisions or actions taken. We also confirm that messages have an impact on the performance of our simulated task., Comment: 4 pages, 3 figures, published as a Late Breaking Report in RO-MAN 2024

The prompt spectra of gamma-ray bursts are known to follow broadband continuum behavior over decades in energy. GRB 221009A, given the moniker the brightest of all time (BOAT), is the brightest gamma-ray burst identified in half a century of observations, and was first identified by the Fermi Gamma-ray Burst Monitor (GBM). On behalf of the Fermi-GBM Team, Lesage et al. (2023) described the initial GBM analysis. Ravasio et al. (2024) report the identification of a spectral line in part of the prompt emission of this burst, which they describe as evolving over 80 s from $\sim$12 MeV to 6 MeV. We report a GBM Team analysis on the Ravasio Line: 1) We cannot identify an instrumental effect that could have produced this signal, and 2) our method of calculating the statistical significance of the line shows it easily exceeds the 5$\sigma$ discovery threshold. We additionally comment on the claim of the line beginning at earlier time intervals, up to 37 MeV, as reported in Zhang et al. (2024). We find that it is reasonable to utilize these measurements for characterization of the line evolution, with caution. We encourage theoretical studies exploring this newly discovered gamma-ray burst spectral feature, unless any rigorous alternative explanation unrelated to the emission from GRB 221009A is identified.

Combined, joint, intra-governmental, inter-agency and multinational (CJIIM) operations require rapid data sharing without the bottlenecks of metadata curation and alignment. Curation and alignment is particularly infeasible for external open source information (OSINF), e.g., social media, which has become increasingly valuable in understanding unfolding situations. Large language models (transformers) facilitate semantic data and metadata alignment but are inefficient in CJIIM settings characterised as denied, degraded, intermittent and low bandwidth (DDIL). Vector symbolic architectures (VSA) support semantic information processing using highly compact binary vectors, typically 1-10k bits, suitable in a DDIL setting. We demonstrate a novel integration of transformer models with VSA, combining the power of the former for semantic matching with the compactness and representational structure of the latter. The approach is illustrated via a proof-of-concept OSINF data discovery portal that allows partners in a CJIIM operation to share data sources with minimal metadata curation and low communications bandwidth. This work was carried out as a bridge between previous low technology readiness level (TRL) research and future higher-TRL technology demonstration and deployment.

Large language model (LLM) chatbots are susceptible to biases and hallucinations, but current evaluations of mental wellness technologies lack comprehensive case studies to evaluate their practical applications. Here, we address this gap by introducing the MHealth-EVAL framework, a new role-play based interactive evaluation method designed specifically for evaluating the appropriateness, trustworthiness, and safety of mental wellness chatbots. We also introduce Psyfy, a new chatbot leveraging LLMs to facilitate transdiagnostic Cognitive Behavioral Therapy (CBT). We demonstrate the MHealth-EVAL framework's utility through a comparative study of two versions of Psyfy against standard baseline chatbots. Our results showed that Psyfy chatbots outperformed the baseline chatbots in delivering appropriate responses, engaging users, and avoiding untrustworthy responses. However, both Psyfy and the baseline chatbots exhibited some limitations, such as providing predominantly US-centric resources. While Psyfy chatbots were able to identify most unsafe situations and avoid giving unsafe responses, they sometimes struggled to recognize subtle harmful intentions when prompted in role play scenarios. Our study demonstrates a practical application of the MHealth-EVAL framework and showcases Psyfy's utility in harnessing LLMs to enhance user engagement and provide flexible and appropriate responses aligned with an evidence-based CBT approach.

We propose that gamma-ray burst pulses are produced when highly-relativistic jets sweep across an observer's line-of-sight. We hypothesize that axisymmetric jet profiles, coupled with special relativistic effects, produce the time-reversed properties of gamma-ray burst pulses. Curvature resulting from rapid jet expansion is responsible for much of the observed pulse asymmetry and hard-to-soft evolution. The relative obliqueness with which the jet crosses the line-of-sight explains the known GRB pulse morphological types. We explore two scenarios: one in which a rigid/semi-rigid jet moves laterally, and the other in which a ballistic jet sprays material from a laterally-moving nozzle. The ballistic jet model is favored based upon its consistency with standard emission mechanisms., Comment: 18 pages, 13 figures, accepted for publication in The Astrophysical Journal

Concept Bottleneck Models (CBMs) are regarded as inherently interpretable because they first predict a set of human-defined concepts which are used to predict a task label. For inherent interpretability to be fully realised, and ensure trust in a model's output, it's desirable for concept predictions to use semantically meaningful input features. For instance, in an image, pixels representing a broken bone should contribute to predicting a fracture. However, current literature suggests that concept predictions often rely on irrelevant input features. We hypothesise that this occurs when dataset labels include inaccurate concept annotations, or the relationship between input features and concepts is unclear. In general, the effect of dataset labelling on concept representations remains an understudied area. In this paper, we demonstrate that CBMs can learn to map concepts to semantically meaningful input features, by utilising datasets with a clear link between the input features and the desired concept predictions. This is achieved, for instance, by ensuring multiple concepts do not always co-occur and, therefore provide a clear training signal for the CBM to distinguish the relevant input features for each concept. We validate our hypothesis on both synthetic and real-world image datasets, and demonstrate under the correct conditions, CBMs can learn to attribute semantically meaningful input features to the correct concept predictions., Comment: Main paper: 8 pages, 9 figures, Appendix: 14 pages, 21 figures. This paper is a preprint

Optical bottle beams, characterized by their unique three-dimensional dark core, have garnered substantial interest due to their potential applications across multiple domains of science and technology. This paper delves into the current methods used to create these beams and provides a method to obscure their nodal planes through coaxial non-interfering orthogonally polarized beams to generate bottle beams with enhanced uniformity. Experimental and theoretical results show the enhanced vector bottle beam maintains a smaller, more spherically uniform potential well and interesting quasi-particle polarization characteristics.

We explore fitting gamma-ray burst spectra with three physically-motivated models, and thus revisit the viability of synchrotron radiation as the primary source of GRB prompt emission. We pick a sample of 100 bright GRBs observed by the Fermi Gamma-ray Burst Monitor (GBM), based on their energy flux values. In addition to the standard empirical spectral models used in previous GBM spectroscopy catalogs, we also consider three physically-motivated models; (a) a Thermal Synchrotron model, (b) a Band model with a High-energy Cutoff, and (c) a Smoothly Broken Power Law (SBPL) model with a Multiplicative Broken Power Law (MBPL). We then adopt the Bayesian information criterion (BIC) to compare the fits obtained and choose the best model. We find that 42% of the GRBs from the fluence spectra and 23% of GRBs from the peak-flux spectra have one of the three physically-motivated models as their preferred one. From the peak-flux spectral fits, we find that the low-energy index distributions from the empirical model fits for long GRBs peak around the synchrotron value of -2/3, while the two low-energy indices from the SBPL+MBPL fits of long GRBs peak close to the -2/3 and -3/2 values expected for a synchrotron spectrum below and above the cooling frequency., Comment: arXiv admin note: text overlap with arXiv:2103.13528

This work reveals an evidential signal that emerges from the uncertainty value in Evidential Deep Learning (EDL). EDL is one example of a class of uncertainty-aware deep learning approaches designed to provide confidence (or epistemic uncertainty) about the current test sample. In particular for computer vision and bidirectional encoder large language models, the `evidential signal' arising from the Dirichlet strength in EDL can, in some cases, discriminate between classes, which is particularly strong when using large language models. We hypothesise that the KL regularisation term causes EDL to couple aleatoric and epistemic uncertainty. In this paper, we empirically investigate the correlations between misclassification and evaluated uncertainty, and show that EDL's `evidential signal' is due to misclassification bias. We critically evaluate EDL with other Dirichlet-based approaches, namely Generative Evidential Neural Networks (EDL-GEN) and Prior Networks, and show theoretically and empirically the differences between these loss functions. We conclude that EDL's coupling of uncertainty arises from these differences due to the use (or lack) of out-of-distribution samples during training.

Accelerated muon beams have been considered for next-generation studies of high-energy lepton-antilepton collisions and neutrino oscillations. However, high-brightness muon beams have not yet been produced. The main challenge for muon acceleration and storage stems from the large phase-space volume occupied by the beam, derived from the muon production mechanism through the decay of pions from proton collisions. Ionization cooling is the technique proposed to decrease the muon beam phase-space volume. Here we demonstrate a clear signal of ionization cooling through the observation of transverse emittance reduction in beams that traverse lithium hydride or liquid hydrogen absorbers in the Muon Ionization Cooling Experiment (MICE). The measurement is well reproduced by the simulation of the experiment and the theoretical model. The results shown here represent a substantial advance towards the realization of muon-based facilities that could operate at the energy and intensity frontiers., Comment: 23 pages and 5 figures

The recent discovery of a kilonova from the long duration gamma-ray burst, GRB 211211A, challenges classification schemes based on temporal information alone. Gamma-ray properties of GRB 211211A reveal an extreme event, which stands out among both short and long GRBs. We find very short variations (few ms) in the lightcurve of GRB 211211A and estimate ~1000 for the Lorentz factor of the outflow. We discuss the relevance of the short variations in identifying similar long GRBs resulting from compact mergers. Our findings indicate that in future gravitational wave follow-up campaigns, some long duration GRBs should be treated as possible strong gravitational wave counterparts., Comment: 10 pages 5 figures, submitted to AAS journals

Concept Bottleneck Models (CBMs) first map raw input(s) to a vector of human-defined concepts, before using this vector to predict a final classification. We might therefore expect CBMs capable of predicting concepts based on distinct regions of an input. In doing so, this would support human interpretation when generating explanations of the model's outputs to visualise input features corresponding to concepts. The contribution of this paper is threefold: Firstly, we expand on existing literature by looking at relevance both from the input to the concept vector, confirming that relevance is distributed among the input features, and from the concept vector to the final classification where, for the most part, the final classification is made using concepts predicted as present. Secondly, we report a quantitative evaluation to measure the distance between the maximum input feature relevance and the ground truth location; we perform this with the techniques, Layer-wise Relevance Propagation (LRP), Integrated Gradients (IG) and a baseline gradient approach, finding LRP has a lower average distance than IG. Thirdly, we propose using the proportion of relevance as a measurement for explaining concept importance., Comment: Accepted into the AAAI-23 workshop Representation Learning for Responsible Human-Centric AI (R2HCAI) as a 4 page paper. This version also includes an additional 47 pages for the appendix and contains additional figures and tables

Working from a model of Gaussian pixel noise, we present and unify over twenty-five years of developments in the statistical analysis of the photon transfer conversion gain measurement. We then study a two-sample estimator of the conversion gain that accounts for the general case of non-negligible dark noise. The moments of this estimator are ill-defined (their integral representations diverge) and so we propose a method for assigning pseudomoments, which are shown to agree with actual sample moments under mild conditions. A definition of optimal sample size pairs for this two-sample estimator is proposed and used to find approximate optimal sample size pairs that allow experimenters to achieve a predetermined measurement uncertainty with as little data as possible. The conditions under which these approximations hold are also discussed. Design and control of experiment procedures are developed and used to optimally estimate a per-pixel conversion gain map of a real image sensor. Experimental results show excellent agreement with theoretical predictions and are backed up with Monte Carlo simulation. The per-pixel conversion gain estimates are then applied in a demonstration of per-pixel read noise estimation of the same image sensor. The results of this work open the door to a comprehensive pixel-level adaptation of the photon transfer method., Comment: 16 pages, 5 figures

Short gamma-ray bursts are associated with binary neutron star mergers, which are multimessenger astronomical events that have been observed both in gravitational waves and in the multiband electromagnetic spectrum. Depending on the masses of the stars in the binary and on details of their largely unknown equation of state, a dynamically evolving and short-lived neutron star may be formed after the merger, existing for approximately 10-300 ms before collapsing to a black hole. Numerical relativity simulations across different groups consistently show broad power spectral features in the 1-5 kHz range in the post-merger gravitational wave signal, which is inaccessible by current gravitational-wave detectors but could be seen by future third generation ground-based detectors in the next decade. This implies the possibility of quasiperiodic modulation of the emitted gamma-rays in a subset of events where a neutron star is formed shortly prior to the final collapse to a black hole. Here we present two such signals identified in the short bursts GRB 910711 and GRB 931101B from archival BATSE data, which are compatible with the predictions from numerical relativity., Comment: 20 pages, 11 figures, 6 tables. Version accepted for publication in Nature