LLMs are increasingly deployed as Artificial Moral Advisors (AMA) in a variety of contexts: what kind of conversational patterns should they display? In this paper, we study how AMA can help their interlocutors "stay with the uncertainty". We propose three modes of uncertainty (Perspective-Multiplying, Tension-Preserving, Process-Reflecting) and compare them against three control conditions (Baseline, Persuasive, Sycophantic). A user-agent LLM engages in a dialogue on an ethical dilemma with an AMA following a specific uncertainty strategy, and completes pre- and post-conversation questionnaires. We further examine the effect of two persona prompt formats (Declarative and Narrative). We found that (1) no single model dominates as a simulated user agent, with open models aligning with human ambiguity through between-persona divergence and closed models through within-persona hedging; (2) declarative personas better capture initial stance diversity while narrative personas show more realistic belief revision; (3) all six AMA strategies produce distinguishable conversational patterns; and (4) uncertainty strategies differ not in how much stance revision they produce, but in the qualit
We investigate the emergence of cohesive flocking in open, boundless space using a multi-agent reinforcement learning framework. Agents integrate positional and orientational information from their closest topological neighbours and learn to balance alignment and attractive interactions by optimizing a local cost function that penalizes both excessive separation and close-range crowding. The resulting Vicsek-like dynamics is robust to algorithmic implementation details and yields cohesive collective motion with high polar order. The optimal policy is dominated by strong aligning interactions when agents are sufficiently close to their neighbours, and a flexible combination of alignment and attraction at larger separations. We further characterize the internal structure and dynamics of the resulting groups using liquid-state metrics and neighbour exchange rates, finding qualitative agreement with empirical observations in starling flocks. These results suggest that flocking may emerge in groups of moving agents as an adaptive response to the biological imperatives of staying together while avoiding collisions.
Visuomotor policies trained via behavior cloning are vulnerable to covariate shift, where small deviations from expert trajectories can compound into failure. Common strategies to mitigate this issue involve expanding the training distribution through human-in-the-loop corrections or synthetic data augmentation. However, these approaches are often labor-intensive, rely on strong task assumptions, or compromise the quality of imitation. We introduce Latent Policy Barrier, a framework for robust visuomotor policy learning. Inspired by Control Barrier Functions, LPB treats the latent embeddings of expert demonstrations as an implicit barrier separating safe, in-distribution states from unsafe, out-of-distribution (OOD) ones. Our approach decouples the role of precise expert imitation and OOD recovery into two separate modules: a base diffusion policy solely on expert data, and a dynamics model trained on both expert and suboptimal policy rollout data. At inference time, the dynamics model predicts future latent states and optimizes them to stay within the expert distribution. Both simulated and real-world experiments show that LPB improves both policy robustness and data efficiency, e
Saliency methods compute heat maps that highlight portions of an input that were most {\em important} for the label assigned to it by a deep net. Evaluations of saliency methods convert this heat map into a new {\em masked input} by retaining the $k$ highest-ranked pixels of the original input and replacing the rest with \textquotedblleft uninformative\textquotedblright\ pixels, and checking if the net's output is mostly unchanged. This is usually seen as an {\em explanation} of the output, but the current paper highlights reasons why this inference of causality may be suspect. Inspired by logic concepts of {\em completeness \& soundness}, it observes that the above type of evaluation focuses on completeness of the explanation, but ignores soundness. New evaluation metrics are introduced to capture both notions, while staying in an {\em intrinsic} framework -- i.e., using the dataset and the net, but no separately trained nets, human evaluations, etc. A simple saliency method is described that matches or outperforms prior methods in the evaluations. Experiments also suggest new intrinsic justifications, based on soundness, for popular heuristic tricks such as TV regularization
Online hate and harassment poses a threat to the digital safety of people globally. In light of this risk, there is a need to equip as many people as possible with advice to stay safer online. We interviewed 24 experts to understand what threats and advice internet users should prioritize to prevent or mitigate harm. As part of this, we asked experts to evaluate 45 pieces of existing hate-and-harassment-specific digital-safety advice to understand why they felt advice was viable or not. We find that experts frequently had competing perspectives for which threats and advice they would prioritize. We synthesize sources of disagreement, while also highlighting the primary threats and advice where experts concurred. Our results inform immediate efforts to protect users from online hate and harassment, as well as more expansive socio-technical efforts to establish enduring safety.
A microscopic, stochastic, minimal model for collective and cohesive motion of identical self-propelled particles is introduced. Even though the particles interact strictly locally in a very noisy manner, we show that cohesion can be maintained, even in the zero-density limit of an arbitrarily large flock in an infinite space. The phase diagram spanned by the two main parameters of our model, which encode the tendencies for particles to align and to stay together, contains non-moving "gas", "liquid"' and "solid" phases separated from their moving counterparts by the onset of collective motion. The "gas/liquid" and "liquid/solid" are shown to be first-order phase transitions in all cases. In the cohesive phases, we study also the diffusive properties of individuals and their relation to the macroscopic motion and to the shape of the flock.
We study thermal behavior of a recently introduced Hartree ensemble approximation, which allows for non-perturbative inhomogeneous field configurations as well as for approximate thermalization, in the $φ^4$ model in 1+1 dimensions. Using ensembles with a free field thermal distribution as out-of-equilibrium initial conditions we determine thermalization time scales. The time scale for which the system stays in approximate quantum thermal equilibrium is an indication of the time scales for which the approximation method stays reasonable. This time scale turns out to be two orders of magnitude larger than the time scale for thermalization, in the range of couplings and temperatures studied. We also discuss simplifications of our method which are numerically more efficient and make a comparison with classical dynamics.
Objective: Predicting length of stay after elective spine surgery is essential for optimizing patient outcomes and hospital resource use. This systematic review synthesizes computational methods used to predict length of stay in this patient population, highlighting model performance and key predictors. Methods: Following PRISMA guidelines, we systematically searched PubMed, Google Scholar, and ACM Digital Library for studies published between December 1st, 2015, and December 1st, 2024. Eligible studies applied statistical or machine learning models to predict length of stay for elective spine surgery patients. Three reviewers independently screened studies and extracted data. Results: Out of 1,263 screened studies, 29 studies met inclusion criteria. Length of stay was predicted as a continuous, binary, or percentile-based outcome. Models included logistic regression, random forest, boosting algorithms, and neural networks. Machine learning models consistently outperformed traditional statistical models, with AUCs ranging from 0.94 to 0.99. K-Nearest Neighbors and Naive Bayes achieved top performance in some studies. Common predictors included age, comorbidities (notably hypertensi
Humanity must solve the problems that threaten human health during long-term space flight. An unexpected solution is suggested by an experiment on people staying in a closed life support system model, conducted more than 50 years ago in Krasnoyarsk. The high level of carbon dioxide established after the start of the experiment changed the metabolism of the inhabitants. We present the interpretation of the results of the Krasnoyarsk experiment through understanding the effect of hypercapnia on metabolism for the first time in this work. Hypercapnia suppressed the possibility of developing metabolic acidosis and activated fat oxidation. The change in the respiratory quotient caused by hypercapnia stabilized the composition of the gas mixture in the closed life support system and ensured the stability of the system.
Tourism significantly affects the economies of many countries. Understanding the causal relationship between the length of overnight stay and traveller's expenditure is crucial for stakeholders to characterize spending profiles and to design marketing strategies. Causal mechanisms differ between personal and work-related travel because the decision-making processes have different drivers and constraints. We apply context-specific independence relations to model causal mechanisms in contexts specified by trip purpose and identify the causal effect of the length of stay on expenditure. Using the international visitor survey data on foreign travellers to Finland, we fit a hierarchical Bayesian model to estimate the posterior distribution of the counterfactual expenditure due to extending the length of stay by one night. We also perform a Bayesian sensitivity analysis of the estimated causal effect with respect to omitted variable bias.
Objective: To describe the association of demographic factors with triage acuity, hospital admission rates, and length of stay (LOS) for adult patients in the emergency department (ED). Methods: We performed a retrospective cross-sectional analysis using publicly available electronic health records describing visits to the ED of a single US medical center during 2011-2019. The primary exposures of interest were self-reported gender, race/ethnicity, and age. The outcomes studied were triage acuity, admission to hospital, and LOS in the ED. Odds ratios were calculated using propensity-score matching. Analyses were adjusted for confounding variables, including vital signs and diagnoses. Key Results: Black patients were more likely than White patients to experience long stays before admission but not before discharge. Men were more likely than women to be triaged as urgent or admitted, and had shorter stays. Patients over 30 were likelier to be triaged as urgent or admitted, but had longer stays.
COVID-19 is a respiratory disease that caused a global pandemic in 2019. It is highly infectious and has the following symptoms: fever or chills, cough, shortness of breath, fatigue, muscle or body aches, headache, the new loss of taste or smell, sore throat, congestion or runny nose, nausea or vomiting, and diarrhea. These symptoms vary in severity; some people with many risk factors have been known to have lengthy hospital stays or die from the disease. In this paper, we analyze patients' electronic health records (EHR) to predict the severity of their COVID-19 infection using the length of stay (LOS) as our measurement of severity. This is an imbalanced classification problem, as many people have a shorter LOS rather than a longer one. To combat this problem, we synthetically create alternate oversampled training data sets. Once we have this oversampled data, we run it through an Artificial Neural Network (ANN), which during training has its hyperparameters tuned using Bayesian optimization. We select the model with the best F1 score and then evaluate it and discuss it.
We study network revenue management problems motivated by applications such as railway ticket sales and hotel room bookings. Requests, each requiring a resource for a consecutive stay, arrive sequentially with known arrival probabilities. We investigate two scenarios: the accept-or-reject scenario, where a request can be fulfilled by assigning any available resource; and the BAM-based scenario, which generalizes the former by incorporating customer preferences through the basic attraction model (BAM), allowing the platform to offer an assortment of available resources from which the customer may choose. We develop polynomial-time policies and evaluate their performance using approximation ratios, defined as the ratio between the expected revenue of our policy and that of the optimal online algorithm. When each arrival has a fixed request type (e.g., the interval of the stay is fixed), we establish constant-factor guarantees: a ratio of 1 - 1/e for the accept-or-reject scenario and 0.25 for the BAM-based scenario. We further extend these results to the case where the request type is random (e.g., the interval of the stay is random). In this setting, the approximation ratios incur an
Using all U.S. Airbnb reservations created in 2019-2024 (booking-count weighted), we quantify pandemic-era shifts in nights per booking (NPB) and the mechanism behind them. The mean rose from 3.68 pre-COVID to 4.36 during restrictions and stabilized near 4.07 post-2021 (about 10% above 2019); the booking-weighted median moved from 2 to 3 nights. A two-parameter log-normal fits best by wide AIC/BIC margins, indicating heavy tails. A negative-binomial model with month effects implies post-vaccine bookings are 6.5% shorter than restriction-era bookings, while pre-COVID bookings are 16% shorter. In a two-part model at 28 nights, the booking share of month-plus stays rose from 1.43% (pre) to 2.72% (restriction) and settled at 2.04% (post); conditional means among long stays were about 55-60 nights. Thus the higher average reflects more long stays rather than longer long stays. A SARIMA(0,1,1)(0,1,1)12 with pandemic-phase dummies improves fit (LR=8.39, df=2, p=0.015), consistent with a structural level shift.
Detecting AI generated images is a challenging yet essential task. A primary difficulty arises from the detectors tendency to rely on spurious patterns, such as compression artifacts, which can influence its decisions. These issues often stem from specific patterns that the detector associates with the real data distribution, making it difficult to isolate the actual generative traces. We argue that an image should be classified as fake if and only if it contains artifacts introduced by the generative model. Based on this premise, we propose Stay Positive, an algorithm designed to constrain the detectors focus to generative artifacts while disregarding those associated with real data. Experimental results demonstrate that detectors trained with Stay Positive exhibit reduced susceptibility to spurious correlations, leading to improved generalization and robustness to post processing. Additionally, unlike detectors that associate artifacts with real images, those that focus purely on fake artifacts are better at detecting inpainted real images.
The paper investigates the escalating concerns surrounding the surge in diabetes cases, exacerbated by the COVID-19 pandemic, and the subsequent strain on medical resources. The research aims to construct a predictive model quantifying factors influencing inpatient hospital stay durations for diabetes patients, offering insights to hospital administrators for improved patient management strategies. The literature review highlights the increasing prevalence of diabetes, emphasizing the need for continued attention and analysis of urban-rural disparities in healthcare access. International studies underscore the financial implications and healthcare burden associated with diabetes-related hospitalizations and complications, emphasizing the significance of effective management strategies. The methodology involves a quantitative approach, utilizing a dataset comprising 10,000 observations of diabetic inpatient encounters in U.S. hospitals from 1999 to 2008. Predictive modeling techniques, particularly Generalized Linear Models (GLM), are employed to develop a model predicting hospital stay durations based on patient demographics, admission types, medical history, and treatment regimen.
Reach-Avoid-Stay (RAS) optimal control enables systems such as robots and air taxis to reach their targets, avoid obstacles, and stay near the target. However, current methods for RAS often struggle with handling complex, dynamic environments and scaling to high-dimensional systems. While reinforcement learning (RL)-based reachability analysis addresses these challenges, it has yet to tackle the RAS problem. In this paper, we propose a two-step deep deterministic policy gradient (DDPG) method to extend RL-based reachability method to solve RAS problems. First, we train a function that characterizes the maximal robust control invariant set within the target set, where the system can safely stay, along with its corresponding policy. Second, we train a function that defines the set of states capable of safely reaching the robust control invariant set, along with its corresponding policy. We prove that this method results in the maximal robust RAS set in the absence of training errors and demonstrate that it enables RAS in complex environments, scales to high-dimensional systems, and achieves higher success rates for the RAS task compared to previous methods, validated through one simu
Predicting hospital length of stay (LoS) stands as a critical factor in shaping public health strategies. This data serves as a cornerstone for governments to discern trends, patterns, and avenues for enhancing healthcare delivery. In this study, we introduce a robust hybrid deep learning model, a combination of Multi-layer Convolutional (CNNs) deep learning, Gated Recurrent Units (GRU), and Dense neural networks, that outperforms 11 conventional and state-of-the-art Machine Learning (ML) and Deep Learning (DL) methodologies in accurately forecasting inpatient hospital stay duration. Our investigation delves into the implementation of this hybrid model, scrutinising variables like geographic indicators tied to caregiving institutions, demographic markers encompassing patient ethnicity, race, and age, as well as medical attributes such as the CCS diagnosis code, APR DRG code, illness severity metrics, and hospital stay duration. Statistical evaluations reveal the pinnacle LoS accuracy achieved by our proposed model (CNN-GRU-DNN), which averages at 89% across a 10-fold cross-validation test, surpassing LSTM, BiLSTM, GRU, and Convolutional Neural Networks (CNNs) by 19%, 18.2%, 18.6%,
We consider random walks conditioned to stay positive. When the mean of increments is zero and variance is finite it is known that they converge to the Rayleigh distribution. In the present paper we derive a Berry-Esseen type estimate and show that the rate of convergence is of order $n^{-1/2}$.