We consider a variant of the multiarmed bandit problem where jobs queue for service, and service rates of different servers may be unknown.

The advancement of Large Language Models (LLMs) enables flexible and interpretable automatic evaluations. In the field of machine translation evaluation, utilizing LLMs with translation error annotations based on Multidimensional Quality Metrics (MQM) yields more human-aligned judgments. However, current LLM-based evaluation methods still face challenges in accurately identifying error spans and assessing their severity. In this paper, we propose HiMATE, a Hierarchical Multi-Agent Framework for Machine Translation Evaluation. We argue that existing approaches inadequately exploit the fine-grained structural and semantic information within the MQM hierarchy. To address this, we develop a hierarchical multi-agent system grounded in the MQM error typology, enabling granular evaluation of subtype errors. Two key strategies are incorporated to further mitigate systemic hallucinations within the framework: the utilization of the model's self-reflection capability and the facilitation of agent discussion involving asymmetric information. Empirically, HiMATE outperforms competitive baselines across different datasets in conducting human-aligned evaluations. Further analyses underscore its significant advantage in error span detection and severity assessment, achieving an average F1-score improvement of 89% over the best-performing baseline. We make our code and data publicly available at https://github.com/nlp2ct-shijie/HiMATE.

Most people spend their lunch breaks grabbing a sandwich or going for a walk. But soon it could be possible to get a full-body MRI scan which detects the earliest stages of cancer during your lunch hour, thanks to AI. Health tech pioneer Ezra has launched its screening service in the UK, marking a major expansion beyond the US. Their AI-powered scans currently last an hour and cover 13 organs, with the added option of an extra lung CT scan and heart disease screening. As cancer rates are rising – especially among young people – the company say they are the best defence against the disease. With early detection, treatment can start earlier and prognosis improves dramatically.

In this study, we utilized statistical analysis and machine learning methods to examine whether rehabilitation exercises can improve patients post-stroke functional abilities, as well as forecast the improvement in functional abilities. Our dataset is patients' rehabilitation exercises and demographic information recorded in the unstructured electronic health records (EHRs) data and free-text rehabilitation procedure notes. We collected data for 265 stroke patients from the University of Pittsburgh Medical Center. We employed a pre-existing natural language processing (NLP) algorithm to extract data on rehabilitation exercises and developed a rule-based NLP algorithm to extract Activity Measure for Post-Acute Care (AM-PAC) scores, covering basic mobility (BM) and applied cognitive (AC) domains, from procedure notes. Changes in AM-PAC scores were classified based on the minimal clinically important difference (MCID), and significance was assessed using Friedman and Wilcoxon tests. To identify impactful exercises, we used Chi-square tests, Fisher's exact tests, and logistic regression for odds ratios. Additionally, we developed five machine learning models-logistic regression (LR), Adaboost (ADB), support vector machine (SVM), gradient boosting (GB), and random forest (RF)-to predict outcomes in functional ability. Statistical analyses revealed significant associations between functional improvements and specific exercises. The RF model achieved the best performance in predicting functional outcomes. In this study, we identified three rehabilitation exercises that significantly contributed to patient post-stroke functional ability improvement in the first two months. Additionally, the successful application of a machine learning model to predict patient-specific functional outcomes underscores the potential for precision rehabilitation.

We consider a variant of the multiarmed bandit problem where jobs queue for service, and service rates of different servers may be unknown. We study algorithms that minimize queue-regret: the (expected) difference between the queue-lengths obtained by the algorithm, and those obtained by a "genie"-aided matching algorithm that knows exact service rates. A naive view of this problem would suggest that queue-regret should grow logarithmically: since queue-regret cannot be larger than classical regret, results for the standard MAB problem give algorithms that ensure queue-regret increases no more than logarithmically in time. Our paper shows surprisingly more complex behavior. In particular, the naive intuition is correct as long as the bandit algorithm's queues have relatively long regenerative cycles: in this case queue-regret is similar to cumulative regret, and scales (essentially) logarithmically. However, we show that this "early stage" of the queueing bandit eventually gives way to a "late stage", where the optimal queue-regret scaling is O(1/t). We demonstrate an algorithm that (order-wise) achieves this asymptotic queue-regret, and also exhibits close to optimal switching time from the early stage to the late stage.

In the rapidly evolving landscape of social media, the introduction of new emojis in Unicode release versions presents a structured opportunity to explore digital language evolution. Analyzing a large dataset of sampled English tweets, we examine how newly released emojis gain traction and evolve in meaning. We find that community size of early adopters and emoji semantics are crucial in determining their popularity. Certain emojis experienced notable shifts in the meanings and sentiment associations during the diffusion process. Additionally, we propose a novel framework utilizing language models to extract words and pre-existing emojis with semantically similar contexts, which enhances interpretation of new emojis. The framework demonstrates its effectiveness in improving sentiment classification performance by substituting unknown new emojis with familiar ones. This study offers a new perspective in understanding how new language units are adopted, adapted, and integrated into the fabric of online communication.

Emotional contagion refers to the sharing of emotional states between individuals, and it has been observed in both animal and human models that the infectivity of negative emotions is much greater than that of positive emotions [1]. Negative emotional contagion has a powerful effect on our relationships - family, friends, teams, etc. - and can lead, for example, to depressive behavior in healthy people who live with depressed individuals. It is urgent to understand the mechanism of emotional contagion, especially negative emotional contagion. Emotional contagion has long been regarded as reflecting a mimicry-based process, for which mimicry of emotional expressions and its consequent feedback function are assumed and can be evoked by higher-order social processes or by a simple emotion-to-action response as well as the primary mimicry-based process [2]. At present, the emotional contagion models mostly adopt behavioral analysis and questionnaires, which are often affected by subjects' subjective factors. They have mainly focused on behavioral experiment such as analysing people's posts containing emotional information to extract affective evidence [3], using the Positive And Negative Affective Schedule (PANAS) scale to measure positive and negative emotions as a quantitive research [4] and the mathematical simulation model of emotional contagion in crowd evacuation [5]. Although behavioral analysis and questionaires can provide valuable insights into emotional contagion, they have limitations in terms of capturing the neural mechanisms, timing, and subtleties of this phenomenon.

Healthcare executives today believe AI will deliver value for the industry faster than previously thought, according to a new survey of senior healthcare executives representing leading hospitals, health plans, life sciences organizations and employers. The third annual Optum Survey on AI in Health Care found that 59% of respondents expect their organizations to see a full return on their AI investments in under three years. That's up 90% since 2018, when only 31% of respondents expected to break even that quickly. The overall anticipated time frame to achieve ROI was 3.6 years in this year's survey, down from 5.3 years in 2018 and 4.7 years in 2019. Confidence in recognizing cost savings from AI appeared to increase as organizations progressed on the maturity curve. Among those who identified themselves as being in the late stages of AI deployment, 57% indicated they'd achieve their ROI in less than two years, as compared to respondents in the early (33%) and mid (26%) stages.

We explored metabolic pathways related to early-stage BCa (Galactose metabolism and Starch and sucrose metabolism) and to late-stage BCa (Glycine, serine, and threonine metabolism, Arginine and proline metabolism, Glycerophospholipid metabolism, and Galactose metabolism) as well as those common to both stages pathways. The central metabolite impacting the most cancerogenic genes (AKT, EGFR, MAPK3) in early stage is d-glucose, while late-stage BCa is characterized by significant fold changes in several metabolites: glycerol, choline, 13(S)-hydroxyoctadecadienoic acid, 2′-fucosyllactose. Insulin was also seen to play an important role in late stages of BCa. The best performing model was able to predict metabolite class with an accuracy of 82.54% and the area under precision-recall curve (PRC) of 0.84 on the training set. The same model was applied to three separate sets of metabolites obtained from public sources, one set of the late-stage metabolites and two sets of the early-stage metabolites.

--Malaria is a life-threatening disease affecting millions. Microscopy-based assessment of thin blood films is a standard method to (i) determine malaria species and (ii) quanti-tate high-parasitemia infections. Full automation of malaria microscopy by machine learning (ML) is a challenging task because field-prepared slides vary widely in quality and presentation, and artifacts often heavily outnumber relatively rare parasites. In this work, we describe a complete, fully-automated framework for thin film malaria analysis that applies ML methods, including convolutional neural nets (CNNs), trained on a large and diverse dataset of field-prepared thin blood films. Quanti-tation and species identification results are close to sufficiently accurate for the concrete needs of drug resistance monitoring and clinical use-cases on field-prepared samples. We focus our methods and our performance metrics on the field use-case requirements. We discuss key issues and important metrics for the application of ML methods to malaria microscopy. Index T erms --malaria, automated microscopy, deep neural networks, gradient boosted trees I. I NTRODUCTION Malaria is a mosquito-borne disease caused by Plasmodium species ( P . Manual microscopy examination of Giemsa-stained blood films is a widespread malaria diagnosis method. Key use-cases include diagnosis; species identification (ID) to guide treatment [2]; and quantitation of parasites for drug resistance studies, to track how fast a drug clears parasites from the blood. However, a lack of training, high inter-sample variability in preparation and presentation, and difficult field conditions can result in poor accuracy [3], [4]. Also, lack of trained personnel limits the number of drug resistance sentinel sites. Malaria microscopy is a difficult task for automated image-processing and machine learning (ML) systems for two reasons: Field-prepared blood films vary widely in quality and presentation; and parasites are small (with feature size close to optical limits of resolution), rare, highly variable, and easily confused with non-parasite objects (artifacts). But it is also a high-value target, due to the potential benefit for so many people, and also because automated systems have some concrete advantages: They can be widely deployed, solving the expert-training bottleneck; they can examine more blood volume per patient, reducing variability in quantitation caused by Poisson statistics; and their results are reproducible.