Growth in the complexity of advanced systems is mirrored by a growth in the number of engineering requirements and related upstream and downstream tasks. These requirements are typically expressed in natural language and require human expertise to manage. Natural language processing (NLP) technology has long been seen as promising to increase requirements engineering (RE) productivity but has yet to demonstrate substantive benefits. The recent addition of large language models (LLMs) to the NLP toolbox is now generating renewed enthusiasm in the hope that it will overcome past shortcomings. This article scrutinizes this claim by reviewing the application of LLMs for engineering requirements tasks. We survey the success of applying LLMs and the scale to which they have been used. We also identify groups of challenges shared across different engineering requirement tasks. These challenges show how this technology has been applied to RE tasks that need reassessment. We finalize by drawing a parallel to other engineering fields with similar challenges and how they have been overcome in the past – and suggest these as future directions to be investigated.

The target backsheath field acceleration mechanism is one of the main mechanisms of laser-driven proton acceleration (LDPA) and strongly depends on the comprehensive performance of the ultrashort ultra-intense lasers used as the driving sources. The successful use of the SG-II Peta-watt (SG-II PW) laser facility for LDPA and its applications in radiographic diagnoses have been manifested by the good performance of the SG-II PW facility. Recently, the SG-II PW laser facility has undergone extensive maintenance and a comprehensive technical upgrade in terms of the seed source, laser contrast and terminal focus. LDPA experiments were performed using the maintained SG-II PW laser beam, and the highest cutoff energy of the proton beam was obviously increased. Accordingly, a double-film target structure was used, and the maximum cutoff energy of the proton beam was up to 70 MeV. These results demonstrate that the comprehensive performance of the SG-II PW laser facility was improved significantly.

Frequent freezing injury greatly influences winter wheat production; thus, effective prevention and a command of agricultural production are vital. The freezing injury monitoring method integrated with ‘3S’ (geographic information systems (GIS), global positioning system (GPS) and remote sensing (RS)) technology has an unparalleled advantage. Using HuanJing (HJ)-1A/1B satellite images of a winter wheat field in Shanxi Province, China plus a field survey, crop types and winter wheat planting area were identified through repeated visual interpretations of image information and spatial analyses conducted in GIS. Six vegetation indices were extracted from processed HJ-1A/1B satellite images to determine whether the winter wheat suffered from freezing injury and its degree of severity and recovery, using change vector analysis (CVA), the freeze injury representative vegetation index and the combination of the two methods, respectively. Accuracy of the freezing damage classification results was verified by determining the impact of freezing damage on yield and quantitative analysis. The CVA and the change of normalized difference vegetation index (ΔNDVI) monitoring results were different so a comprehensive analysis of the combination of CVA and ΔNDVI was performed. The area with serious freezing injury covered 0.9% of the total study area, followed by the area of no freezing injury (3.5%), moderate freezing injury (10.2%) and light freezing injury (85.4%). Of the moderate and serious freezing injury areas, 0.2% did not recover; 1.2% of the no freezing injury and light freezing injury areas showed optimal recovery, 15.6% of the light freezing injury and moderate freezing injury areas showed poor recovery, and the remaining areas exhibited general recovery.

In the coastal ocean, interactions of waves and currents with large roughness elements, similar in size to wave orbital excursions, generate drag and dissipate energy. These boundary layer dynamics differ significantly from well-studied small-scale roughness. To address this problem, we derived spatially and phase-averaged momentum equations for combined wave–current flows over rough bottoms, including the canopy layer containing obstacles. These equations were decomposed into steady and oscillatory parts to investigate the effects of waves on currents, and currents on waves. We applied this framework to analyse large-eddy simulations of combined oscillatory and steady flows over hemisphere arrays (diameter $D$), in which current ($U_c$), wave velocity ($U_w$) and period ($T$) were varied. In the steady momentum budget, waves increase drag on the current, and this is balanced by the total stress at the canopy top. Dispersive stresses from oscillatory flow around obstacles are increasingly important as $U_w/U_c$ increases. In the oscillatory momentum budget, acceleration in the canopy is balanced by pressure gradient, added-mass and form drag forces; stress gradients are small compared to other terms. Form drag is increasingly important as the Keulegan–Carpenter number $KC=U_wT/D$ and $U_c/U_w$ increase. Decomposing the drag term illustrates that a quadratic relationship predicts the observed dependences of steady and oscillatory drag on $U_c/U_w$ and $KC$. For large roughness elements, bottom friction is well represented by a friction factor ($f_w$) defined using combined wave and current velocities in the canopy layer, which is proportional to drag coefficient and frontal area per unit plan area, and increases with $KC$ and $U_c/U_w$.

A classical result for the simple symmetric random walk with 2n steps is that the number of steps above the origin, the time of the last visit to the origin, and the time of the maximum height all have exactly the same distribution and converge when scaled to the arcsine law. Motivated by applications in genomics, we study the distributions of these statistics for the non-Markovian random walk generated from the ascents and descents of a uniform random permutation and a Mallows(q) permutation and show that they have the same asymptotic distributions as for the simple random walk. We also give an unexpected conjecture, along with numerical evidence and a partial proof in special cases, for the result that the number of steps above the origin by step 2n for the uniform permutation generated walk has exactly the same discrete arcsine distribution as for the simple random walk, even though the other statistics for these walks have very different laws. We also give explicit error bounds to the limit theorems using Stein’s method for the arcsine distribution, as well as functional central limit theorems and a strong embedding of the Mallows(q) permutation which is of independent interest.

Loot boxes provide randomized rewards in video games; their purchase is linked to disordered gambling and they are present in approximately half of UK video games. The relative novelty of loot boxes means that regulators and policymakers in various jurisdictions are still deciding how to regulate them. The People's Republic of China (PRC) is the first, and presently only, jurisdiction to legally require companies to disclose the probabilities of obtaining randomized loot box rewards – an approach that is also favored by the industry as self-regulation. This study is the first to assess paid loot box prevalence in the PRC and companies’ discretionary interpretations of probability disclosure regulations. Loot boxes were found in 91 of the 100 highest-grossing PRC iPhone games. Of games deemed suitable for children aged 12+, 90.5% contained loot boxes. Probability disclosures could not be found for 4.4% of games containing loot boxes. Disclosures were implemented through various methods both in-game and on the games’ official websites; however, consistent with the concept of ‘sludge,’ only 5.5% used the most prominent format of automatically displaying the probabilities on the in-game loot box purchase page. Loot box probability disclosures should be uniform and visually prominent to best help inform consumers.

Object To know the treatment experiences of patients with depression in general hospital, and their suggestion on the present mental health services. To provides theoretical support for optimizing the mental health services model in general hospital. Methods Qualitative interviews were given to 10 patients diagnosed as depressive disorder from department of psychiatry, Renmin Hospital of Wuhan University. Questions were asked from following aspects: the process of seeking medical help before hospitalization in psychiatry department(when, where, who, referral), evaluation of drug treatment (efficacy, side effects, doctors) and psychotherapy, satisfaction and suggestions on medical services(including hospital environment, medical staff and etc.), social rehabilitation (stigma, social functional recovery, survival / quality of life etc.), socioeconomic factors (health insurance coverage, financial resources, social support resources). Answers referred to above factors were conducted to statistical analysis. Results showed departments of neurology, gastroenterology, endocrinology, Chinese traditional medicine, obstetrics and gynecology are the first medical choice of depressive patients; the average number of departments before they get psychiatric treatment is five, the average number of referral is 2. Medication side effects are the most concerned problems. Enough and reasonable information given from doctor about drug's name, efficacy, side effects, and length of treatment can improve patient's compliance, even can predict recurrence rate. Discrimination medical personnel can reduce compliance of patients: contempt, pessimism, avoidance, activity limitations, and impolite language and behavior. Psychotherapy can shorten the hospital stay. Health insurance coverage, job, social support can predict social functional recovery. Conclusions Doctors in general hospital should improve their ability in recognizing depression to ensure referral patients timely; Education of depression, medication-related knowledge should offered by psychiatrists to improve patients’ treatment compliance, and reduce relapse rate; psychological therapy should be incorporated into psychiatric conventional treatment; Medical personnel should be conscious of using positive affirmation, respectful words and actions; effective communication with families is necessary for promoting patients’ social rehabilitation.

Let $\overline{t}(n)$ be the number of overpartitions in which (i) the difference between successive parts may be odd only if the larger part is overlined and (ii) if the smallest part is odd then it is overlined. Ramanujan-type congruences for $\overline{t}(n)$ modulo small powers of $2$ and $3$ have been established. We present two infinite families of congruences modulo $5$ and $27$ for $\overline{t}(n)$, the first of which generalises a recent result of Chern and Hao [‘Congruences for two restricted overpartitions’, Proc. Math. Sci. 129 (2019), Article 31].

We use the results of a supernova light-curve population synthesis to predict the range of possible supernova light curves arising from a population of single-star progenitors that lead to type IIP supernovae. We calculate multiple models varying the initial mass, explosion energy, nickel mass and nickel mixing and then compare these to type IIP supernovae with detailed light curve data and pre-explosion imaging progenitor constraints. Where a good fit is obtained to observations, we are able to achieve initial progenitor and nickel mass estimates from the supernova lightcurve that are comparable in precision to those obtained from progenitor imaging. For 2 of the 11 IIP supernovae considered our fits are poor, indicating that more progenitor models should be included in our synthesis or that our assumptions, regarding factors such as stellar mass loss rates or the rapid final stages of stellar evolution, may need to be revisited in certain cases. Using the results of our analysis we are able to show that most of the type IIP supernovae have an explosion energy of the order of log(Eexp/ergs) = 50.52 ± 0.10 and that both the amount of nickel in the supernovae and the amount of mixing may have a dependence on initial progenitor mass.

We present results of a supernova lightcurve population synthesis, predicting the range of possible supernova lightcurves arising from a population of progenitor stars that include interacting binary systems. We show that the known diversity of supernova lightcurves can be interpreted as arising from binary interactions. Given detailed models of the progenitor stars, we are able to the determine what parameters within these stars determine the shape of their supernova lightcurve. The primary factors are the mass of supernova ejecta and the mass of hydrogen in the final progenitor. We find that there is a continuum of lightcurve behaviour from type IIP, IIL, to IIb supernovae related to the range of hydrogen and ejecta masses. Most type IIb supernovae arise from a relatively narrow range of initial masses from 10 to 15 M⊙. We also find a few distinct lightcurves that are the result of stellar mergers.

Air pollution is a persistent and well-established public health problem in India: emissions from coal-fired power plants have been associated with over 80,000 premature deaths in 2015. Premature deaths could rise by four to five times this number by 2050 without additional pollution controls. We site a model 500 MW coal-fired electricity generating unit at eight locations in India and examine the benefits and costs of retrofitting the plant with a flue-gas desulfurization unit to reduce sulfur dioxide emissions. We quantify the mortality benefits associated with the reduction in sulfates (fine particles) and value these benefits using estimates of the value per statistical life transferred to India from high income countries. The net benefits of scrubbing vary widely by location, reflecting differences in the size of the exposed population. They are highest at locations in the densely populated north of India, which are also among the poorest states in the country.

In several lately published studies, the association between single-nucleotide polymorphism (SNP, rs12252) of IFITM3 and the risk of influenza is inconsistent. To further understand the association between the SNP of IFITM3 and the risk of influenza, we searched related studies in five databases including PubMed published earlier than 9 November 2017. Ten sets of data from nine studies were included and data were analysed by Revman 5.0 and Stata 12.0 in our updated meta-analysis, which represented 1365 patients and 5425 no-influenza controls from four different ethnicities. Here strong association between rs12252 and influenza was found in all four genetic models. The significant differences in the allelic model (C vs. T: odds ratio (OR) = 1.35, 95% confidence interval (CI) (1.03–1.79), P = 0.03) and homozygote model (CC vs. TT: OR = 10.63, 95% CI (3.39–33.33), P < 0.00001) in the Caucasian subgroup were discovered, which is very novel and striking. Also novel discoveries were found in the allelic model (C vs. T: OR = 1.37, 95% CI (1.08–1.73), P = 0.009), dominant model (CC + CT vs. TT: OR = 1.48, 95% CI (1.08–2.02), P = 0.01) and homozygote model (CC vs. TT: OR = 2.84, 95% CI (1.36–5.92), P = 0.005) when we compared patients with mild influenza with healthy individuals. Our meta-analysis suggests that single-nucleotide T to C polymorphism of IFITM3 associated with increasingly risk of severe and mild influenza in both Asian and Caucasian populations.

The response of soil microbial communities to soil quality changes is a sensitive indicator of soil ecosystem health. The current work investigated soil microbial communities under different fertilization treatments in a 31-year experiment using the phospholipid fatty acid (PLFA) profile method. The experiment consisted of five fertilization treatments: without fertilizer input (CK), chemical fertilizer alone (MF), rice (Oryza sativa L.) straw residue and chemical fertilizer (RF), low manure rate and chemical fertilizer (LOM), and high manure rate and chemical fertilizer (HOM). Soil samples were collected from the plough layer and results indicated that the content of PLFAs were increased in all fertilization treatments compared with the control. The iC15:0 fatty acids increased significantly in MF treatment but decreased in RF, LOM and HOM, while aC15:0 fatty acids increased in these three treatments. Principal component (PC) analysis was conducted to determine factors defining soil microbial community structure using the 21 PLFAs detected in all treatments: the first and second PCs explained 89.8% of the total variance. All unsaturated and cyclopropyl PLFAs except C12:0 and C15:0 were highly weighted on the first PC. The first and second PC also explained 87.1% of the total variance among all fertilization treatments. There was no difference in the first and second PC between RF and HOM treatments. The results indicated that long-term combined application of straw residue or organic manure with chemical fertilizer practices improved soil microbial community structure more than the mineral fertilizer treatment in double-cropped paddy fields in Southern China.

Considering as many as 70% of massive stars interact with a binary companion (Sana et al.2012, 2014), we created a new model of the optical nebular emission of HII regions by combining the results of the Binary Population and Spectral Synthesis (BPASS, Eldridge, Stanway et al.2017) code with the photoionization code (CLOUDY). This is discussed more in detail in Xiao et al.2018a. Then we use this model to explore a variety of emission-line diagnostics of CCSN host HII regions from the PMAS/PPAK Integral-field Supernova hosts COmpilation (PISCO, Galbany et al.2018). We determine the age, metallicity and gas parameters for H II regions associated with CCSNe, contrasting the above variables to distribution type II and type Ibc SNe. We find their nebular emission and CCSN progenitor types are largely determined by past and ongoing binary interactions, for example mass loss, mass gain and stellar mergers. However we note these two types SNe have little preference in their host environment metallicity measured by oxygen abundance or in progenitor initial mass, except that at lower metallicities supernovae are more likely to be of type II. The BPASS and nebular emission models are available from bpass.auckland.ac.nz and warwick.ac.uk/bpass.