Comparing Fine-Tuning, Zero and Few-Shot Strategies with Large Language Models in Hate Speech Detection in English

Large Language Models(LLMs)are increasingly demonstrating their ability to understand natural language and solve complex tasks,especially through text generation.One of the relevant capabilities is contextual learning,which involves the ability to receive instructions in natural language or task demonstrations to generate expected outputs for test instances without the need for additional training or gradient updates.In recent years,the popularity of social networking has provided a medium through which some users can engage in offensive and harmful online behavior.In this study,we investigate the ability of different LLMs,ranging from zero-shot and few-shot learning to fine-tuning.Our experiments show that LLMs can identify sexist and hateful online texts using zero-shot and few-shot approaches through information retrieval.Furthermore,it is found that the encoder-decoder model called Zephyr achieves the best results with the fine-tuning approach,scoring 86.811%on the Explainable Detection of Online Sexism(EDOS)test-set and 57.453%on the Multilingual Detection of Hate Speech Against Immigrants and Women in Twitter(HatEval)test-set.Finally,it is confirmed that the evaluated models perform well in hate text detection,as they beat the best result in the HatEval task leaderboard.The error analysis shows that contextual learning had difficulty distinguishing between types of hate speech and figurative language.However,the fine-tuned approach tends to produce many false positives.

SIL1 improves cognitive impairment in APP23/PS45 mice by regulating amyloid precursor protein processing and Aβ generation

SIL1,an endoplasmic reticulum(ER)-resident protein,is reported to play a protective role in Alzheimer’s disease(AD).However,the effect of SIL1 on amyloid precursor protein(APP)processing remains unclear.In this study,the role of SIL1 in APP processing was explored both in vitro and in vivo.In the in vitro experiment,SIL1 was either overexpressed or knocked down in cells stably expressing the human Swedish mutant APP695.In the in vivo experiment,AAV-SIL1-EGFP or AAV-EGFP was microinjected into APP23/PS45 mice and their wild-type littermates.Western blotting(WB),immunohistochemistry,RNA sequencing(RNA-seq),and behavioral experiments were performed to evaluate the relevant parameters.Results indicated that SIL1 expression decreased in APP23/PS45 mice.Overexpression of SIL1 significantly decreased the protein levels of APP,presenilin-1(PS1),and C-terminal fragments(CTFs)of APP in vivo and in vitro.Conversely,knockdown of SIL1 increased the protein levels of APP,β-site APP cleavage enzyme 1(BACE1),PS1,and CTFs,as well as APP mRNA expression in 2EB2 cells.Furthermore,SIL1 overexpression reduced the number of senile plaques in APP23/PS45 mice.Importantly,Y-maze and Morris Water maze tests demonstrated that SIL1 overexpression improved cognitive impairment in APP23/PS45 mice.These findings indicate that SIL1 improves cognitive impairment in APP23/PS45 mice by inhibiting APP amyloidogenic processing and suggest that SIL1 is a potential therapeutic target for AD by modulating APP processing.

Automatic depression recognition by intelligent speech signal processing:A systematic survey

Depression has become one of the most common mental illnesses in the world.For better prediction and diagnosis,methods of automatic depression recognition based on speech signal are constantly proposed and updated,with a transition from the early traditional methods based on hand‐crafted features to the application of architectures of deep learning.This paper systematically and precisely outlines the most prominent and up‐to‐date research of automatic depression recognition by intelligent speech signal processing so far.Furthermore,methods for acoustic feature extraction,algorithms for classification and regression,as well as end to end deep models are investigated and analysed.Finally,general trends are summarised and key unresolved issues are identified to be considered in future studies of automatic speech depression recognition.

贝托司汀手性中间体(S)-CPMA整细胞催化合成工艺研究

针对不同条件对贝托司汀手性中间体(S)-CPMA的整细胞催化合成影响,进行单因素实验和正交设计优化研究.首先从耐热克鲁维酵母突变株SXBP-02预处理方式、不同催化反应介质、pH、底物质量浓度、温度、时间、催化剂用量、补料次数等8个方面进行考察,揭示其中pH、底物质量浓度、补料次数和催化剂用量4个因素对(S)-CPMA催化效率的影响较大;进一步对其进行正交优化,得出最佳工艺条件:在pH8的PEG4000双水相体系中,底物质量浓度为6g/L、补料次数为4次、催化剂用量为3g/L,温度为40℃条件下反应36h,可得到(S)-CPMA产率为88.1%,对映选择性在98.5%以上,4个因素对(S)-CPMA收率的影响主次顺序依次为补料次数>pH>催化剂用量>底物质量浓度,其中补料次数、pH为显著性影响因素,与优化前相比,底物质量浓度提高3倍,产率提高了6%.

基于D-S证据理论改进AHP-熵权的流域洪涝灾害评估研究

考虑致灾因子危险性、孕灾环境敏感性以及承灾体易损性,选取指标构建小清河流域洪涝灾害风险评估指标体系,提出一种基于D-S证据理论的改进AHP-熵权法计算指标权重,求取洪涝灾害风险指数,运用自然断点分级法确定洪涝灾害风险等级,分析小清河流域洪涝灾害风险空间分布情况。结果表明:小清河流域洪涝灾害风险总体上表现出南低北高的趋势,其中高风险区和较高风险区分别占流域面积的8.7%和14.3%,主要分布在小清河干流以及主要支流两岸。所得评估结果同“利奇马”台风发生期间实际洪灾风险分布情况一致,对比证明基于D-S证据理论的改进AHP-熵权法优于AHP和熵权法,可为小清河流域防洪减灾决策提供依据。

Improved Ant Lion Optimizer with Deep Learning Driven Arabic Hate Speech Detection

Arabic is the world’s first language,categorized by its rich and complicated grammatical formats.Furthermore,the Arabic morphology can be perplexing because nearly 10,000 roots and 900 patterns were the basis for verbs and nouns.The Arabic language consists of distinct variations utilized in a community and particular situations.Social media sites are a medium for expressing opinions and social phenomena like racism,hatred,offensive language,and all kinds of verbal violence.Such conduct does not impact particular nations,communities,or groups only,extending beyond such areas into people’s everyday lives.This study introduces an Improved Ant Lion Optimizer with Deep Learning Dirven Offensive and Hate Speech Detection(IALODL-OHSD)on Arabic Cross-Corpora.The presented IALODL-OHSD model mainly aims to detect and classify offensive/hate speech expressed on social media.In the IALODL-OHSD model,a threestage process is performed,namely pre-processing,word embedding,and classification.Primarily,data pre-processing is performed to transform the Arabic social media text into a useful format.In addition,the word2vec word embedding process is utilized to produce word embeddings.The attentionbased cascaded long short-term memory(ACLSTM)model is utilized for the classification process.Finally,the IALO algorithm is exploited as a hyperparameter optimizer to boost classifier results.To illustrate a brief result analysis of the IALODL-OHSD model,a detailed set of simulations were performed.The extensive comparison study portrayed the enhanced performance of the IALODL-OHSD model over other approaches.

An Optimal Method for Speech Recognition Based on Neural Network

Natural language processing technologies have become more widely available in recent years,making them more useful in everyday situations.Machine learning systems that employ accessible datasets and corporate work to serve the whole spectrum of problems addressed in computational linguistics have lately yielded a number of promising breakthroughs.These methods were particularly advantageous for regional languages,as they were provided with cut-ting-edge language processing tools as soon as the requisite corporate information was generated.The bulk of modern people are unconcerned about the importance of reading.Reading aloud,on the other hand,is an effective technique for nour-ishing feelings as well as a necessary skill in the learning process.This paper pro-posed a novel approach for speech recognition based on neural networks.The attention mechanism isfirst utilized to determine the speech accuracy andfluency assessments,with the spectrum map as the feature extraction input.To increase phoneme identification accuracy,reading precision,for example,employs a new type of deep speech.It makes use of the exportchapter tool,which provides a corpus,as well as the TensorFlow framework in the experimental setting.The experimentalfindings reveal that the suggested model can more effectively assess spoken speech accuracy and readingfluency than the old model,and its evalua-tion model’s score outcomes are more accurate.

AN ANALYSIS OF ACOUSTIC CHARACTERISTICS OFCLEFT PALATE SPEECH WITH COMPUTERIZED SPEECH SIGNAL PROCESSING SYSTEM

The acoustic characteristics or the chinese vowels of 24 children with cleft palate and 10 normal control children were analyzed by computerized speech signal processing system (CSSPS),and the speech articulation was judged with Glossary of clert palate speech(GCPS).The listening judgement showed that the speech articulation was significantly different between the two groups(P<0.01).The objective quantitative measurement suggested that the formant pattern(FP)of vowels in children with cleft palate was different from that of normal control children except vowel[a](P< 0.05).The acoustic vowelgraph or the Chinese vowels which demonstrated directly the relationship of vocal space and speech perception was stated with the first formant frequence(F1)and the second formant frequence(F2).The authors conclude that the values or F1 and F2 point out the upward and backward tongue movement to close the clert, which reflects the vocal characteristics of trausmission of clert palate speech.

Noise Removal in Speech Processing Using Spectral Subtraction

Spectral subtraction is used in this research as a method to remove noise from noisy speech signals in the frequency domain. This method consists of computing the spectrum of the noisy speech using the Fast Fourier Transform (FFT) and subtracting the average magnitude of the noise spectrum from the noisy speech spectrum. We applied spectral subtraction to the speech signal “Real graph”. A digital audio recorder system embedded in a personal computer was used to sample the speech signal “Real graph” to which we digitally added vacuum cleaner noise. The noise removal algorithm was implemented using Matlab software by storing the noisy speech data into Hanning time-widowed half-overlapped data buffers, computing the corresponding spectrums using the FFT, removing the noise from the noisy speech, and reconstructing the speech back into the time domain using the inverse Fast Fourier Transform (IFFT). The performance of the algorithm was evaluated by calculating the Speech to Noise Ratio (SNR). Frame averaging was introduced as an optional technique that could improve the SNR. Seventeen different configurations with various lengths of the Hanning time windows, various degrees of data buffers overlapping, and various numbers of frames to be averaged were investigated in view of improving the SNR. Results showed that using one-fourth overlapped data buffers with 128 points Hanning windows and no frames averaging leads to the best performance in removing noise from the noisy speech.

Improved Attentive Recurrent Network for Applied Linguistics-Based Offensive Speech Detection

Applied linguistics is one of the fields in the linguistics domain and deals with the practical applications of the language studies such as speech processing,language teaching,translation and speech therapy.The ever-growing Online Social Networks(OSNs)experience a vital issue to confront,i.e.,hate speech.Amongst the OSN-oriented security problems,the usage of offensive language is the most important threat that is prevalently found across the Internet.Based on the group targeted,the offensive language varies in terms of adult content,hate speech,racism,cyberbullying,abuse,trolling and profanity.Amongst these,hate speech is the most intimidating form of using offensive language in which the targeted groups or individuals are intimidated with the intent of creating harm,social chaos or violence.Machine Learning(ML)techniques have recently been applied to recognize hate speech-related content.The current research article introduces a Grasshopper Optimization with an Attentive Recurrent Network for Offensive Speech Detection(GOARN-OSD)model for social media.The GOARNOSD technique integrates the concepts of DL and metaheuristic algorithms for detecting hate speech.In the presented GOARN-OSD technique,the primary stage involves the data pre-processing and word embedding processes.Then,this study utilizes the Attentive Recurrent Network(ARN)model for hate speech recognition and classification.At last,the Grasshopper Optimization Algorithm(GOA)is exploited as a hyperparameter optimizer to boost the performance of the hate speech recognition process.To depict the promising performance of the proposed GOARN-OSD method,a widespread experimental analysis was conducted.The comparison study outcomes demonstrate the superior performance of the proposed GOARN-OSD model over other state-of-the-art approaches.

An Intervention Study of Language Cognition and Emotional Speech Community Method for Children’s Speech Disorders

Speech disorders are a common type of childhood disease.Through experimental intervention,this study aims to improve the vocabulary comprehension levels and language ability of children with speech disorders through the language cognition and emotional speech community method.We also conduct a statistical analysis of the inter-ventional effect.Among children with speech disorders in Dongguan City,224 were selected and grouped accord-ing to their receptive language ability and IQ.The 112 children in the experimental group(EG)received speech therapy with language cognitive and emotional speech community,while the 112 children in the control group(CG)only received conventional treatment.After six months of experimental intervention,the Peabody Picture Vocabulary Test-Revised(PPVT-R)was used to test the language ability of the two groups.Overall,we employed a quantitative approach to obtain numerical values,examine the variables identified,and test hypotheses.Further-more,we used descriptive statistics to explore the research questions related to the study and statistically describe the overall distribution of the demographic variables.The statistical t-test was used to analyze the data.The data shows that after intervention through language cognition and emotional speech community therapy,the PPVT-R score of the EG was significantly higher than that of the CG.Therefore,we conclude that there is a significant difference in language ability between the EG and CG after the therapy.Although both groups improved,the post-therapy language level of EG is significantly higher than that of CG.The total effective rate in EG is higher than CG,and the difference is statistically significant(p<0.05).Therefore,we conclude that the language cogni-tion and emotional speech community method is effective as an interventional treatment of children’s speech dis-orders and that it is more effective than traditional treatment methods.

Audio-visual keyword transformer for unconstrained sentence-level keyword spotting

As one of the most effective methods to improve the accuracy and robustness of speech tasks,the audio-visual fusion approach has recently been introduced into the field of Keyword Spotting(KWS).However,existing audio-visual keyword spotting models are limited to detecting isolated words,while keyword spotting for unconstrained speech is still a challenging problem.To this end,an Audio-Visual Keyword Transformer(AVKT)network is proposed to spot keywords in unconstrained video clips.The authors present a transformer classifier with learnable CLS tokens to extract distinctive keyword features from the variable-length audio and visual inputs.The outputs of audio and visual branches are combined in a decision fusion module.As humans can easily notice whether a keyword appears in a sentence or not,our AVKT network can detect whether a video clip with a spoken sentence contains a pre-specified keyword.Moreover,the position of the keyword is localised in the attention map without additional position labels.Exper-imental results on the LRS2-KWS dataset and our newly collected PKU-KWS dataset show that the accuracy of AVKT exceeded 99%in clean scenes and 85%in extremely noisy conditions.The code is available at https://github.com/jialeren/AVKT.

Strategies for Improving the Effectiveness of Professional Practice for Full-Time Professional Master Degree Postgraduate in Mineral Processing Engineering

In order to gain practical experience and hands-on skills,full-time professional master degree postgraduate in mineral processing engineering should engage in professional practices.Nonetheless,a series of problems,including insufficient time for practice,low management level,inadequate implementation of the double-supervisor system,and poor results of professional practice,has reduced the effectiveness of professional practice.In view of the aforementioned problems and the characteristics of the discipline,this paper proposes several strategies for improving the effectiveness of professional practice for postgraduates in mineral processing engineering.

2022年四川泸定M_(S)6.8地震震源破裂过程及强地面运动模拟

首先,利用区域宽频带波形拟合,反演了2022年泸定M_(S)6.8地震序列主震和部分余震的震源机制解和震源深度;其次,基于有限断层模型反演方法,使用区域宽频带波形数据反演了此次泸定地震震源破裂过程,并根据得到的震源破裂模型计算了峰值地面速度(PGV)分布。结果显示,此次地震矩心深度为6.0 km,是一个典型的高角度左旋走滑地震。震源破裂传播方向主要沿断层走向约165°向东南方向传播,由深部震源起始破裂点向浅部扩展,并破裂到地表,地表破裂主要分布在磨西到猛虎岗一带,长度约16 km。震源破裂过程持续时间约20 s,能量主要集中在前15 s内释放。地震释放的标量地震矩为1.07×10^(19)N·m,约等于矩震级M_(W)6.62。地震主体破裂发生在3~6 s之间,最大滑动量达到1.8 m,位于震中东南方向深度约10 km处。除此之外,在震中西北方向深度11 km处和震中东南方向深度18 km处分别发生两个次级破裂,滑动量大约在0.6~1.0 m。主体破裂的破裂长度约20 km,两个次级破裂的长度分别约为4 km和8 km。使用该震源破裂模型计算得到了PGV分布,PGV分布以震中为中心沿断层走向两侧扩散,其长轴与地震断层走向一致,呈NW向,极震区PGV为200~360 cm/s,地震烈度区内受灾严重的村镇均位于极震区。

A Multi-Band Speech Enhancement Algorithm Exploiting Iterative Processing for Enhancement of Single Channel Speech

This paper proposes a multi-band speech enhancement algorithm exploiting iterative processing for enhancement of single channel speech. In the proposed algorithm, the output of the multi-band spectral subtraction (MBSS) algorithm is used as the input signal again for next iteration process. As after the first MBSS processing step, the additive noise transforms to the remnant noise, the remnant noise needs to be further re-estimated. The proposed algorithm reduces the remnant musical noise further by iterating the enhanced output signal to the input again and performing the operation repeatedly. The newly estimated remnant noise is further used to process the next MBSS step. This procedure is iterated a small number of times. The proposed algorithm estimates noise in each iteration and spectral over-subtraction is executed independently in each band. The experiments are conducted for various types of noises. The performance of the proposed enhancement algorithm is evaluated for various types of noises at different level of SNRs using, 1) objective quality measures: signal-to-noise ratio (SNR), segmental SNR, perceptual evaluation of speech quality (PESQ);and 2) subjective quality measure: mean opinion score (MOS). The results of proposed enhancement algorithm are compared with the popular MBSS algorithm. Experimental results as well as the objective and subjective quality measurement test results confirm that the enhanced speech obtained from the proposed algorithm is more pleasant to listeners than speech enhanced by classical MBSS algorithm.

基于外部气相沉积的S+C+L波段低色散斜率大有效面积非零色散位移光纤的设计与制备

针对现有光纤无法满足宽带光密集波分复用系统传输和S+C+L波段粗波分复用的要求,设计了一种具有中心凹陷的三角形芯+环形的折射率剖面,利用外部气相沉积工艺制备了一种非零色散位移光纤,并通过调整第一芯层的相对折射率和第二芯层与第一芯层的半径比,探究了其对光纤衰减、色散斜率和有效面积等参数的影响。研究发现,当第一芯层的相对折射率逐渐增大且第二芯层与第一芯层半径比逐渐减小时,零色散波长和有效面积逐渐减小。当第一芯层的相对折射率在0.52%~0.53%,芯层半径比在2.6~2.7时,光纤的有效面积接近70μm^(2),零色散波长在1420 nm附近,在1550 nm波段的色散系数大于8 ps·nm^(-1)·km^(-1),色散斜率为0.059 ps·nm^(-2)·km^(-1),可以较好地抑制传输过程中光非线性效应,满足长途干线网与城域网的使用要求。

Digital Disparities:How Artificial Intelligence Can Facilitate Anti-Black Racism in the U.S.Healthcare Sector

This paper delves into the intricate interplay between artificial intelligence(AI)systems and the perpetuation of Anti-Black racism within the United States medical industry.Despite the promising potential of AI to enhance healthcare outcomes and reduce disparities,there is a growing concern that these technologies may inadvertently/advertently exacerbate existing racial inequalities.Focusing specifically on the experiences of Black patients,this research investigates how the following AI components:medical algorithms,machine learning,and natural learning processes are contributing to the unequal distribution of medical resources,diagnosis,and health care treatment of those classified as Black.Furthermore,this review employs a multidisciplinary approach,combining insights from computer science,medical ethics,and social justice theory to analyze the mechanisms through which AI systems may encode and reinforce racial biases.By dissecting the three primary components of AI,this paper aims to present a clear understanding of how these technologies work,how they intersect,and how they may inherently perpetuate harmful stereotypes resulting in negligent outcomes for Black patients.Furthermore,this paper explores the ethical implications of deploying AI in healthcare settings and calls for increased transparency,accountability,and diversity in the development and implementation of these technologies.Finally,it is important that I prefer the following paper with a clear and concise definition of what I refer to as Anti-Black racism throughout the text.Therefore,I assert the following:Anti-Black racism refers to prejudice,discrimination,or antagonism directed against individuals or communities of African descent based on their race.It involves the belief in the inherent superiority of one race over another and the systemic and institutional practices that perpetuate inequality and disadvantage for Black people.Furthermore,I proclaim that this form of racism can be manifested in various ways,such as unequal access to opportunities,resources,education,employment,and fair treatment within social,economic,and political systems.It is also pertinent to acknowledge that Anti-Black racism is deeply rooted in historical and societal structures throughout the U.S.borders and beyond,leading to systemic disadvantages and disparities that impact the well-being and life chances of Black individuals and communities.Addressing Anti-Black racism involves recognizing and challenging both individual attitudes and systemic structures that contribute to discrimination and inequality.Efforts to combat Anti-Black racism include promoting awareness,education,advocacy for policy changes,and fostering a culture of inclusivity and equality.

A Historical Review of the Development of Women’s Higher Education in China (1908-2020): Stages, Explanations, and Trends

The development of women’s higher education in China can be divided into four stages:emergence(1908-1948);foundation(1949-1976);accelerating development(1977-2008);and the qualitative leap(2009-2020).This work considers the principal institutional mechanisms that contributed to this development.First,flexibly planned parenthood gradually promoted gender equality and openness in society facilitated by systematic“awards,grants,and loans”initiatives to support women’s higher education economically.Second,compulsory education ensured that left-out and migrant children had access to higher education.Third,effective connectivity across different education types bridged education gaps between those with different levels of education.Fourth,China made great efforts to invite and integrate international experiences that promoted the development of women’s higher education.Looking beyond these achievements,we also discuss the future trends of women’s higher education in China.

浅析国家药监局加入药品检查合作计划(PIC/S)中广东药监面临的形势与发展展望

国家药监局加入药品检查合作计划(PIC/S)的进程已经进入正式申请阶段,这将加速推动我国药品监管体系国际化现代化发展。本文简要介绍了PIC/S的相关情况,梳理了国家药监局加入该组织的主要进程以及国内近年来在药品GMP检查体系和相关能力建设方面的工作进展,展望了该形势下广东省药监局联合港澳地区协同推进药品监管能力建设、凝聚自身实践经验、助力国家药品监管和药品产业发展的工作脉络。

级间分离S形分离钢索高速分离敏感因素分析

级间分离S形分离钢索结构复杂,高速分离过程涉及大变形、复杂非线性接触等问题,其动态变形及力学特性分析是工程设计的难点之一。基于Abaqus建立S形分离钢索动态分离仿真模型,分析了S形分离钢索高速分离动态变形过程及力学特性,并通过地面试验进一步验证建模方法的正确性。研究了S形分离钢索长度、截面直径、拖尾长度、分离速度等多种敏感因素的影响,对工程实际具有一定的指导意义。