Autoregressive video generation relies on history context for content consistency and storytelling. As video histories grow longer, efficiently encoding them remains an open problem - particularly for personal users and local workflows where compute and memory budgets are limited. We present a lightweight history encoder that maps long video histories into short-length embeddings, pretrained with a frame query objective that learns to attend to content features at arbitrary temporal positions. The pretraining stage provides the encoder with dense history coverage on large-scale video data; the subsequent finetuning stage adapts the pretrained encoder under an autoregressive video generation objective to establish content-level consistency. In this way, the lightweight embeddings achieve comparable performance to heavier alternatives. We evaluate the framework with ablative settings and discuss the architecture designs.
Measurements of jet substructure are key to probing the energy frontier at colliders, and many of them use track-based observables which take advantage of the angular precision of tracking detectors. Theoretical calculations of track-based observables require ‘track functions’, which characterize the transverse momentum fraction rq carried by charged hadrons from a fragmenting quark or gluon. This letter presents a direct measurement of rq distributions in dijet events from the 140 fb−1 of proton–proton collisions at s=13 TeV recorded with the ATLAS detector. The data are corrected for detector effects using machine-learning methods. The scale evolution of the moments of the rq distribution is sensitive to non-linear renormalization group evolution equations of QCD, and is compared with analytic predictions. When incorporated into future theoretical calculations, these results will enable a precision program of theory-data comparison for track-based jet substructure observables.
AbstractThis article offers a critical rereading of the historiography on the role of women in the Italian Resistance. It starts with the postwar period, marked by a general silence and the prevailing image of women as mothers and staffette. In the 1970s, the first historical elaboration of women’s experiences began in all northern regions, leading to the now iconic concept of the ‘silent Resistance’. In the 1990s, a dialogue developed with other historiographical categories, such as the concept of ‘civil resistance’ developed by Jacques Sémelin and the ‘war on civilians’, but this approach ran the risk of reducing women’s contribution to ‘powerless’ acts. Although today women’s history is fully integrated into the narrative canon of the Resistance, it faces new challenges, such as the confrontation with ‘other’ (mainly non-European) resistances and new public uses of history. The article suggests that women’s history has been, if not the only, then certainly the most important means by which new dimensions of the partisan movement and the Second World War have been brought to the fore, shedding light on the specificities of the conflict experienced by women, but also shaping the very notion of resistance by overcoming a purely militarist vision.
Se-eun Yoon, Ahmad Bin Rabiah, Zaid Alibadi
et al.
Customers reach out to online live chat agents with various intents, such as asking about product details or requesting a return. In this paper, we propose the problem of predicting user intent from browsing history and address it through a two-stage approach. The first stage classifies a user's browsing history into high-level intent categories. Here, we represent each browsing history as a text sequence of page attributes and use the ground-truth class labels to fine-tune pretrained Transformers. The second stage provides a large language model (LLM) with the browsing history and predicted intent class to generate fine-grained intents. For automatic evaluation, we use a separate LLM to judge the similarity between generated and ground-truth intents, which closely aligns with human judgments. Our two-stage approach yields significant performance gains compared to generating intents without the classification stage.
Thousands of users consult digital archives daily, but the information they can access is unrepresentative of the diversity of documentary history. The sequence-to-sequence architecture typically used for optical character recognition (OCR) - which jointly learns a vision and language model - is poorly extensible to low-resource document collections, as learning a language-vision model requires extensive labeled sequences and compute. This study models OCR as a character level image retrieval problem, using a contrastively trained vision encoder. Because the model only learns characters' visual features, it is more sample efficient and extensible than existing architectures, enabling accurate OCR in settings where existing solutions fail. Crucially, the model opens new avenues for community engagement in making digital history more representative of documentary history.
We argue that observations of the reionization history can be used as a probe of primordial density fluctuations, particularly on small scales. Although the primordial curvature perturbations are well constrained from measurements of cosmic microwave background (CMB) anisotropies and large-scale structure, these observational data probe the curvature perturbations only on large scales, and hence its information on smaller scales will give us further insight on primordial fluctuations. Since the formation of early galaxies is sensitive to the amplitude of small-scale perturbations, and then, in turn, gives an impact on the reionization history, one can probe the primordial power spectrum on small scales through observations of reionization. In this work, we focus on the running spectral indices of the primordial power spectrum to characterize the small-scale perturbations, and investigate their impact on the reionization history using the numerical code \texttt{21cmFAST}, which adopts a simple but commonly used reionization model. We also derive the constraints on the running spectral indices from observations of the reionization history indicated by the luminosity function of the Lyman-$α$ emitters. We show that the reionization history, in combination with large-scale observations such as CMB, would be a useful tool to investigate primordial density fluctuations.
Various processes can be modelled as quasi-reaction systems of stochastic differential equations, such as cell differentiation and disease spreading. Since the underlying data of particle interactions, such as reactions between proteins or contacts between people, are typically unobserved, statistical inference of the parameters driving these systems is developed from concentration data measuring each unit in the system over time. While observing the continuous time process at a time scale as fine as possible should in theory help with parameter estimation, the existing Local Linear Approximation (LLA) methods fail in this case, due to numerical instability caused by small changes of the system at successive time points. On the other hand, one may be able to reconstruct the underlying unobserved interactions from the observed count data. Motivated by this, we first formalise the latent event history model underlying the observed count process. We then propose a computationally efficient Expectation-Maximation algorithm for parameter estimation, with an extended Kalman filtering procedure for the prediction of the latent states. A simulation study shows the performance of the proposed method and highlights the settings where it is particularly advantageous compared to the existing LLA approaches. Finally, we present an illustration of the methodology on the spreading of the COVID-19 pandemic in Italy.
Il presente contributo si propone di far nuova luce sulla storia di un importante ciclo ad affresco del ‘600 fiorentino, raffigurante in 15 lunette i Misteri del Rosario e realizzato dal pittore valdarnese Giovanni da San Giovanni per l’ex monastero di Annalena a Firenze. Il ciclo è stato a lungo dimenticato perché considerato distrutto a inizio XIX secolo. Le pitture murali, invece, vennero risparmiate dalle demolizioni del complesso: riportate su tela dal restauratore emiliano Giovanni Rizzoli, presero la strada della Gran Bretagna alla fine dell’800. Le inedite informazioni sono emerse dallo studio di numerose fonti a stampa e dalla consultazione di diversi archivi in Italia e in Inghilterra. La riscoperta della riproduzione a stampa dei Misteri (pubblicata nel 1904) potrà sicuramente favorire il ritrovamento del ciclo in terra inglese.
The paper aims to shed new light on the history of an important 17th century Florentine fresco cycle, depicting the Mysteries of the Rosary in 15 lunettes, made by Giovanni da San Giovanni in the former Annalena monastery in Florence. The cycle was long forgotten, since it was considered destroyed at the beginning of the 19th century: the artworks survived, however, transferred to canvas by the Emilian restorer Giovanni Rizzoli, and ended up in Great Britain by the end of the 19th century. Many unpublished information emerged from the study of numerous printed sources and from the consultation of numerous archives in Italy and England. The rediscovery of the engravings of the Mysteries (published in 1904) will certainly favor the possible discovery of the cycle in England as well.
Conversational question generation (CQG) serves as a vital task for machines to assist humans, such as interactive reading comprehension, through conversations. Compared to traditional single-turn question generation (SQG), CQG is more challenging in the sense that the generated question is required not only to be meaningful, but also to align with the occurred conversation history. While previous studies mainly focus on how to model the flow and alignment of the conversation, there has been no thorough study to date on which parts of the context and history are necessary for the model. We argue that shortening the context and history is crucial as it can help the model to optimise more on the conversational alignment property. To this end, we propose CoHS-CQG, a two-stage CQG framework, which adopts a CoHS module to shorten the context and history of the input. In particular, CoHS selects contiguous sentences and history turns according to their relevance scores by a top-p strategy. Our model achieves state-of-the-art performances on CoQA in both the answer-aware and answer-unaware settings.
Abstract Background The spread of knowledge on the important implications of a diagnosis of genetic disease does not correspond to a sharing of the knowledge and equal rights of children. Main body It is estimated that about 5% of newborns may have a rare disease that in some cases, if diagnosed early, could have specific treatments that may be able to modify the natural history of the disease. However, in most countries the diagnosis during the first hours of life is limited to a few diseases, due to the high costs and time required for genetic investigations with classical methods. Recently, experimental projects to subject all newborns to a complete DNA analysis, with Next Generation Sequencing techniques, to detect any genetic pathologies as early as possible, have been reported in some countries. The late diagnosis of some genetic diseases that have treatment plans, such as spinal muscular atrophy, can be a serious damage, for anyone who has seen and accompanied the life of a child with this disease and his/her family, before and after, the recent availability of therapies which, if started very early, can lead to an almost normal life. Rapid sequencing and genetic diagnosis are a crucial part of directing inpatient management and this resource should be accessible not only to academic medical centers but also in community settings. Conclusions It is time for a profound reflection that places in Italy, as in other countries, the use of genetic tests in neonatal and pediatric age based on principles of evidence, ethics, and democracy and on clear national guidelines, which also consider organizational aspects.
Objectives: To describe long-COVID symptoms among older adults and to assess the risk factors for two common long-COVID symptoms: fatigue and dyspnea. Methods: This is a multicenter, prospective cohort study conducted in Israel, Switzerland, Spain, and Italy. Individuals were included at least 30 days after their COVID-19 diagnosis. We compared long-COVID symptoms between elderly (aged >65 years) and younger individuals (aged 18-65 years) and conducted univariate and multivariable analyses for the predictors of long-COVID fatigue and dyspnea. Results: A total of 2333 individuals were evaluated at an average of 5 months (146 days [95% confidence interval 142-150]) after COVID-19 onset. The mean age was 51 years, and 20.5% were aged >65 years. Older adults were more likely to be symptomatic, with the most common symptoms being fatigue (38%) and dyspnea (30%); they were more likely to complain of cough and arthralgia and have abnormal chest imaging and pulmonary function tests. Independent risk factors for long-COVID fatigue and dyspnea included female gender, obesity, and closer proximity to COVID-19 diagnosis; older age was not an independent predictor. Conclusion: Older individuals with long-COVID have different persisting symptoms, with more pronounced pulmonary impairment. Women and individuals with obesity are at risk. Further research is warranted to investigate the natural history of long-COVID among the elderly population and to assess possible interventions aimed at promoting rehabilitation and well-being.
Rosagemma Ciliberti, Francesca Lantieri, Rosario Barranco
et al.
The objectives of this study were to obtain information on medical students’ attitudes toward COVID-19 vaccination and to identify the main barriers to its acceptance. We conducted an anonymous online survey on a sample of undergraduate medical students from one main Italian University. The questions were aimed at exploring their attitudes toward vaccination to prevent COVID-19, their perceptions of the risk/threat of COVID-19 and the factors associated with their attitudes toward COVID-19 vaccination. A high percentage of students in our sample stated that they had been vaccinated or that they intended to be vaccinated against the COVID-19 coronavirus. A total of 239 questionnaires were analyzed. Age, social, geographic and demographic characteristics, health conditions and interest in vaccination were recorded; 93% of the students declared that they encouraged vaccination and 83% stated that the reason was “Moral responsibility towards the community”. Four students had not yet been vaccinated, mainly because of “Contradictory information on efficacy and safety”. The Likert-type questions revealed high agreement on the importance of vaccination and whether it should be made mandatory (“indispensable tool” and “ethical duty” were cited to explain this position). The results show a high level of acceptance of COVID-19 vaccination among these medical undergraduates who, being halfway through their training and involved in clinical practice, are already in possession of specific scientific knowledge and, to a small extent, come from different areas of Italy.
Damian Clarke, Manuel Llorca Jaña, Daniel Pailañir
Quantile regression and quantile treatment effect methods are powerful econometric tools for considering economic impacts of events or variables of interest beyond the mean. The use of quantile methods allows for an examination of impacts of some independent variable over the entire distribution of continuous dependent variables. Measurement in many quantative settings in economic history have as a key input continuous outcome variables of interest. Among many other cases, human height and demographics, economic growth, earnings and wages, and crop production are generally recorded as continuous measures, and are collected and studied by economic historians. In this paper we describe and discuss the broad utility of quantile regression for use in research in economic history, review recent quantitive literature in the field, and provide an illustrative example of the use of these methods based on 20,000 records of human height measured across 50-plus years in the 19th and 20th centuries. We suggest that there is considerably more room in the literature on economic history to convincingly and productively apply quantile regression methods.
Jan Zapletal, Raphael Watschinger, Günther Of
et al.
The presented paper concentrates on the boundary element method (BEM) for the heat equation in three spatial dimensions. In particular, we deal with tensor product space-time meshes allowing for quadrature schemes analytic in time and numerical in space. The spatial integrals can be treated by standard BEM techniques known from three dimensional stationary problems. The contribution of the paper is twofold. First, we provide temporal antiderivatives of the heat kernel necessary for the assembly of BEM matrices and the evaluation of the representation formula. Secondly, the presented approach has been implemented in a publicly available library besthea allowing researchers to reuse the formulae and BEM routines straightaway. The results are validated by numerical experiments in an HPC environment.
The author sees the purpose of this study in introducing to the scientific circulation the materials published by Soviet journalists and diplomats based on the results of their stay in fascist Italy. This topic is of particular relevance due to the fact that modern people’s ideas about the cultures of other countries are often superficial and are influenced by stereotypes. The ideological confrontation between Rome and Moscow of the time left a special imprint on these texts: criticizing the fascist regime, the authors of the publications portrayed the Italian people as their victims. The essays and reports considered in the text of the work were written by direct witnesses of the events that took place on the Apennines in the 1920-1930s, which makes them important sources of information regarding the history of Italy and Soviet-Italian relations. The memoirs conceived by the First Plenipotentiary of Soviet Russia in Italy V. Vorovsky, could have been an outstanding historical document, but his premature death thwarted these plans. The first journal publication following a visit by the Soviet correspondent to Italy was a report on the visit of the destroyers Nezamozhnik and Petrovsky to Naples in 1925. Boris Zilpert’s essay on the Italian press was published in the «Journalist» in 1926. It was written based on the conversations in Rome with colleagues. The materials of A. Keen, D. Ilimsky, I. Robin are of considerable value for the purposes of historic and cultural studies. S. Ignatiev in his writings was able to illuminate the situation in the Italian colony of Eritrea. In the 1920s Soviet domestic press also actively collaborated with Italian journalists (G. Giogo, C. Rossi). Almost all materials of the Soviet press of the period under review emphasized the alienness of the fascist regime to the national character of the Italian people. Authors focused their attention on the Nazis’ violation of civil rights and freedoms, the imposition of militarism, and the persecution of communists. The Za rubezhom magazine posted translations of the materials of the foreign press that were written in the similar vein. The author comes to the conclusion that the fascist regime of Benito Mussolini was criticized in the Soviet press from humanistic positions.
The article examines the peculiarities of the formation of the image of the enemy in the textbook of Andrei Shestakov "A Short Course in the History of the USSR." Also, the historiography of the issue is analyzed. It applies to both the Soviet era and the present. Despite the modest attention to this topic by foreign experts, the works that directly affect the issue are highlighted. The main changes in the then system of school education, which led to its unification and formed the requirements for the history lesson in general and the need to develop a textbook in particular, are outlined. The role of Andrei Shestakov, who was one of the first to develop an "ideal" history textbook for the Soviet government, is revealed. His career growth and work with Marxist-Leninist ideology are shown, which in turn helped to achieve this goal.
The process of modification of negative connotations concerning those forces against which the Bolsheviks fought is traced. Thus considering the period of ancient history, the author criticized rich people. The negative image deepened when it came to religion in the Middle Ages. Priests and monks, compared to the wealthy, were perceived not as something "foreign" but more negative as something "hostile." Wealthy peasants received a special color, the term "kulaks" was used for them. The closer A. Shestakov approached the twentieth century in his presentation of historical material, the clearer the formation of the image of the enemy, not only internal but also external. Thus, the first was personified by all the forces against which the Bolsheviks fought. To define such "enemies" used the definition - "counter-revolutionary". The second category was represented by the Entente and the Nazis, who came to power primarily in Germany and Italy.
History (General) and history of Europe, Philosophy. Psychology. Religion
Aim. To evaluate clinical characteristics and perinatal outcomes in a heterogeneous population of Caucasians born in Italy and High Migration Pressure Countries (HMPC) women with GDM living in Piedmont, North Italy. Methods. We retrospectively analyzed data from 586 women referring to our unit (2015–2020). Epidemiological (age and country of origin) and clinical-metabolic features (height, weight, family history of DM, parity, previous history of GDM, OGTT results, and GDM treatment) were collected. The database of certificates of care at delivery was consulted in relation to neonatal/maternal complications (rates of caesarean sections, APGAR score, fetal malformations, and neonatal anthropometry). Results. 43.2% of women came from HMPC; they were younger p<0.0001 and required insulin treatment more frequently than Caucasian women born in Italy (χ2 = 17.8, p=0.007). Higher fasting and 120-minute OGTT levels and gestational BMI increased the risk of insulin treatment (OGTT T0: OR = 1.04, CI 95% 1.016–1.060, p=0.005; OGTT T120: OR = 1.01, CI 95% 1.002–1.020, p=0.02; BMI: OR = 1.089, CI 95% 1.051–1.129, p<0.0001). Moreover, two or more diagnostic OGTT glucose levels doubled the risk of insulin therapy (OR = 2.03, IC 95% 1.145–3.612, p=0.016). We did not find any association between ethnicities and neonatal/maternal complications. Conclusions. In our multiethnic GDM population, the need for intensive care and insulin treatment is high in HPMC women although the frequency of adverse peripartum and newborn outcomes does not vary among ethnic groups. The need for insulin therapy should be related to different genetic backgrounds, dietary habits, and Nutrition Transition phenomena. Thus, nutritional intervention and insulin treatment need to be tailored.
Diseases of the endocrine glands. Clinical endocrinology
In evolutionary algorithms, genetic operators iteratively generate new offspring which constitute a potentially valuable set of search history. To boost the performance of crossover in real-coded genetic algorithm (RCGA), in this paper we propose to exploit the search history cached so far in an online style during the iteration. Specifically, survivor individuals over past few generations are collected and stored in the archive to form the search history. We introduce a simple yet effective crossover model driven by the search history (abbreviated as SHX). In particular, the search history is clustered and each cluster is assigned a score for SHX. In essence, the proposed SHX is a data-driven method which exploits the search history to perform offspring selection after the offspring generation. Since no additional fitness evaluations are needed, SHX is favorable for the tasks with limited budget or expensive fitness evaluations. We experimentally verify the effectiveness of SHX over 4 benchmark functions. Quantitative results show that our SHX can significantly enhance the performance of RCGA, in terms of accuracy.