Target Concept Research Articles

Conceptual Mappings of Metaphorical Euphemisms in Ekegusii Circumcision Ceremonies

Conceptual mappings are a key tenet of the Conceptual Metaphor Theory (CMT) and are sets of systematic correspondences between the source and target domains. As observed in the relevant literature, it is argued that metaphors establish a conceptual link between a source and target concept such that the target domain is understood through a source domain. As a result, the correspondences are constrained by different mapping scopes which help avoid the transfer of just any kind of feature from the source to the target concept. In this study, we sought to examine the conceptual mappings of metaphorical euphemisms in Ekegusii Circumcision Ceremonies. The study analyzed the conceptual mappings of metaphorical euphemisms used in EC based on the essentials of cognitive metaphor analysis and principles of Conceptual Metaphor Theory (CMT). The metaphor-related words (MRWs) identified using the Metaphor Identification Procedure (MIPVU) were subjected to conceptual analysis using the CMT. The conceptualization of the target domain euphemisms was obtained through projections in the source domain such as objects, animals, plants, and human beings. In this view, the study identified the following four generic metaphors in Ekegusii Circumcision Discourse; Circumcision is a Human being, Circumcision is an Animal, Circumcision is a Plant and Circumcision is an Object. The study employed a descriptive research design. A combined method of elicitation and the native speaker’s intuition was used to collect culture-specific metaphorical euphemisms through Focus Group Discussions (FGDs). The quantitative data on the other hand was analyzed using the statistical Package for Social Sciences which is; Chi-square at a significant level of 0.05 was employed to test the association between the overall cross-tabulated corpus of the generic level metaphors versus the influence of demographic variables on the usage of euphemisms in Ekegusii Circumcision ceremonies. The findings of this study not only contribute to the existing literature in cognitive linguistics but also indicate that Ekegusii speakers conceptualize circumcision through metaphorical euphemisms. The findings of this study are of benefit to the ethnographers and metaphor theorists for further reference.

A Cognitive -Semantic Study of Love Metaphor in Mawlawi Poems

This study is entitled "A Cognitive Semantic Study of Love Metaphor in Mawlawi Poems". It is an attempt to construe the semantic aspects of Mawlawi's poetic texts in light of the cognitive semantic approach. The study, through a linguistic and cognitive analysis, clarifies the way meaning is constructed in the selected poems. It adopts the theoretical framework proposed by the cognitive semantic theory of metaphor. That is, according to the cognitive theory of metaphor, it explores the metaphorization of LOVE in the selected poems. In cognitive approach to metaphor, an abstract concept named the target domain is understood in terms of another called the source domain (usually concrete). The source domain is a conceptual domain used metaphorically so as to understand the abstract target concept, i.e., the target domain is a conceptual domain interpreted through the source domain. The current study investigates the ways in which LOVE as an abstract target concept is construed in terms of other source conceptual domains. That is, the primary aim of the study is to scrutinize how LOVE is metaphorized in the poems, what are the source domains the poet relies upon so as to interpret the target domain of LOVE. In other words, what are the more concrete conceptual source domains that are metaphorically used in order to comprehend the abstract target concept of LOVE? Accordingly, the recent study is mainly concerned with conceptualization (the way LOVE is conceptualized) which is the cognitive processes of meaning-making. This means that, it illuminates meaning construction process in Mawlawi literature. The study first provides a brief introduction to Cognitive Linguistics and then the nature of meaning in cognitive semantic approach is highlighted. Following this, the cognitive approach to metaphor is presented. The last section is devoted to elucidating the metaphorical structuring of LOVE in Mawlawi's poems based on the contemporary theory of metaphor.

Метафорическая концептуализация экономических отношений в публицистическом дискурсе финансово-экономической нестабильности (на материалах англоязычных публикаций, отражающих кризис потолка госдолга в США в 2023 г.)

The article examines the metaphorical conceptualization of a crisis economic situation in mass media discourse. The article looks at the metaphorical representation of economic relationships in the period of financial uncertainty caused by the inflation and the debt ceiling crisis in the USA in 2023. The aim of the research is to determine and describe the most relevant metaphorical models depicting the economic phenomena in the given historical period. The scientific novelty lies in describing the ways of metaphorical conceptualization of economic relationships in mass media discourse brought to life by new empirical data of the debt ceiling crisis. The research findings include describing the most relevant metaphorical models that depict the economic phenomena in the articles of The New York Times, The Washington Post, The Economist; determining the nominative field of the target concept ECONOMY; extending the corpus of metaphors from crisis discourse and identifying the primary functions of the metaphors used in the texts under the study.

Agents Targeting the Bacterial Cell Wall as Tools to Combat Gram-Positive Pathogens.

The cell wall is an indispensable element of bacterial cells and a long-known target of many antibiotics. Penicillin, the first discovered beta-lactam antibiotic inhibiting the synthesis of cell walls, was successfully used to cure many bacterial infections. Unfortunately, pathogens eventually developed resistance to it. This started an arms race, and while novel beta-lactams, either natural or (semi)synthetic, were discovered, soon upon their application, bacteria were developing resistance. Currently, we are facing the threat of losing the race since more and more multidrug-resistant (MDR) pathogens are emerging. Therefore, there is an urgent need for developing novel approaches to combat MDR bacteria. The cell wall is a reasonable candidate for a target as it differentiates not only bacterial and human cells but also has a specific composition unique to various groups of bacteria. This ensures the safety and specificity of novel antibacterial agents that target this structure. Due to the shortage of low-molecular-weight candidates for novel antibiotics, attention was focused on peptides and proteins that possess antibacterial activity. Here, we describe proteinaceous agents of various origins that target bacterial cell wall, including bacteriocins and phage and bacterial lysins, as alternatives to classic antibiotic candidates for antimicrobial drugs. Moreover, advancements in protein chemistry and engineering currently allow for the production of stable, specific, and effective drugs. Finally, we introduce the concept of selective targeting of dangerous pathogens, exemplified by staphylococci, by agents specifically disrupting their cell walls.

Exploring the Development of Photoresists in Lacquer under the Background of Global Carbon Neutrality: A Mini Review

With the establishment of a global carbon-neutral target, green and sustainable design concepts, materials, and processes have become the key to scientific and technological innovation, which highlights the development and utilization of environmentally-friendly materials and renewable energy sources. At present, almost all the raw materials in commercial photoresist are from oil resources. In order to implement the policy of “green chemistry”, there is an urgent need to take the place of petroleum with renewable, biodegradable, and green biomass resources in photoresist production. As a kind of green and natural high polymer material from China, lacquer has excellent physical and chemical properties and has great application potential in the field of electronics. In this paper, the potential connections between photoresist and lacquer, especially in the green development of photoresist, are sorted out. The introduction of lacquer and its derivatives may help to improve the environmentally-friendly production of photoresist and its performance. Therefore, in the context of global carbon neutrality, the study of photoresists in the “lacquer” is of great significance in promoting the application of eco-friendly materials in the field of microelectronics.

Novel WRN Helicase Inhibitors Selectively Target Microsatellite-Unstable Cancer Cells.

Microsatellite-unstable (MSI) cancers require WRN helicase to resolve replication stress due to expanded DNA (TA)n dinucleotide repeats. WRN is a promising synthetic lethal target for MSI tumors, and WRN inhibitors are in development. In this study, we used CRISPR-Cas9 base editing to map WRN residues critical for MSI cells, validating the helicase domain as the primary drug target. Fragment-based screening led to the development of potent and highly selective WRN helicase covalent inhibitors. These compounds selectively suppressed MSI model growth in vitro and in vivo by mimicking WRN loss, inducing DNA double-strand breaks at expanded TA repeats and DNA damage. Assessment of biomarkers in preclinical models linked TA-repeat expansions and mismatch repair alterations to compound activity. Efficacy was confirmed in immunotherapy-resistant organoids and patient-derived xenograft models. The discovery of potent, selective covalent WRN inhibitors provides proof of concept for synthetic lethal targeting of WRN in MSI cancer and tools to dissect WRN biology. Significance: We report the discovery and characterization of potent, selective WRN helicase inhibitors for MSI cancer treatment, with biomarker analysis and evaluation of efficacy in vivo and in immunotherapy-refractory preclinical models. These findings pave the way to translate WRN inhibition into MSI cancer therapies and provide tools to investigate WRN biology. See related commentary by Wainberg, p. 1369.

Conceptual Metonymy of COVID-19 Pandemic in Vietnam’s Online Newspapers

Conceptual metonymy which is based on the contingent relationship between a source concept and a target concept helps people perceive the world. The article discusses the concept and principles of conceptual metonymy and analyses the conceptual metonymy of COVID-19 pandemic used in official online newspapers published in Vietnam. 100 articles were selected for analysis. The study focuses on the "PLACE FOR EVENT" metonymy with the aim of shedding light on the Vietnamese’s cognitive ways of thinking. Vietnamese people tend to use the geographical location where the COVID-19 pandemic occurs as a substitute for the COVID-19 pandemic.

Fostering Learning from Errors—Computer-Based Adaptivity at the Transition Between Problem Solving and Explicit Instruction

When learners acquire new content by working on a problem-solving task prior to explicit instruction, their attempts to solve the problem usually represent only partial steps on the way to the target concept. Both, theoretical assumptions on conceptual change as well as empirical findings on effective instructional formats with incorrect solutions, suggest that it is beneficial to address incorrect student solutions in a (subsequent) instruction phase by comparing incorrect and correct solutions. There is initial evidence that learning is most successful when learner compare the correct solution to an incorrect solution that reflects the learners’ conceptual understanding from the problem-solving phase. In the present study, we investigated in a highly controlled experimental design the relevance of this fit between the learners’ individual solution type from the problem-solving phase and the incorrect solution type in the instruction phase for learning success. In a computer-based learning environment, sixth graders worked on a problem-solving task to compare fractions. In the subsequent instruction phase, students in three conditions were given 1) an adaptive comparison, 2) a contra-adaptive comparison, 3) only the correct solution. Overall, there were no differences across conditions regarding the learning success. Further exploratory analyses revealed that only learners with an intermediate prior knowledge benefited from the adaptivity. This finding can be interpreted as indicator that our short intervention only induces conceptual change when basic knowledge is already available.

Maximizing the Effectiveness of Anti-Smoking Campaigns With Targeted Framing Strategies: Evidence From Behavioral and Neurological Studies

Background The global prevalence of smoking remains a significant public health concern, necessitating the development of effective anti-smoking campaigns to promote smoking cessation and discourage smoking initiation. Understanding the factors that influence individual responses to anti-smoking messages is critical for tailoring campaigns to different populations. Focus of the Article This study substantially exhibits the concepts of audience segmentation and targeting, informed by theory and formative research, and monitoring and evaluation research in the context of anti-smoking campaigns. Research Question The article aims to answer how smoking behavior and message framing influence individuals’ counterarguing responses to anti-smoking advertisements and the implications for ad effectiveness. Approach The research investigates the role of smoking behavior (smokers vs. non-smokers) and message framing (positive vs. negative) in shaping individuals’ counterarguing responses to anti-smoking advertisements, using two studies involving behavioral and neuroscientific methods. Methods Study 1 employed an online video task with 1920 participants, examining the effects of smoking behavior and message framing on ad effectiveness. Study 2 used electroencephalography (EEG) to record brain activity from 31 participants while they viewed anti-smoking video advertisements, focusing on the neural correlates of counterarguing and their effect on anti-smoking ad effectiveness. Results Study 1 found that smokers report lower ad effectiveness compared to non-smokers, and observed an interaction between smoking behavior and message framing. Study 2 found that smoking behavior significantly influenced brain activity in the dorsolateral prefrontal cortex (dlPFC), ventrolateral prefrontal cortex (VLPFC), and anterior cingulate cortex (ACC) regions in response to anti-smoking ads. Additionally, a significant interaction was observed between smoking behavior and ad framing on brain activity in all three regions. Recommendations for Research or Practice The findings suggest that anti-smoking campaigns should be tailored to the specific needs and attitudes of smokers and non-smokers, with separate campaigns for each group. Future research could explore the role of other individual differences in shaping counterarguing responses, the long-term effects of exposure to different types of anti-smoking messages on smoking behavior, and using more advanced neuroimaging techniques to investigate the neural processes that underlie counterarguing responses to anti-smoking messages for better flow.

Abstract B020: Novel WRN helicase inhibitors selectively target microsatellite unstable cancer cells

Abstract Microsatellite-unstable (MSI) cancers require WRN helicase to resolve replication stress due to expanded DNA (TA)n-dinucleotide repeats. WRN is a promising synthetic lethal target for MSI tumours, and WRN inhibitors are in development. Here, we used CRISPR-Cas9 base editing to map WRN residues critical for MSI cells, validating the helicase domain as the primary drug target. Fragment-based screening led to the development of potent and highly selective WRN helicase covalent inhibitors. These compounds selectively suppressed MSI model growth In vitro and In vivo by mimicking WRN loss, inducing DNA double-strand breaks at expanded TA-repeats and DNA damage. Assessment of biomarkers in preclinical models linked TA-repeat expansions and mismatch repair (MMR) alterations to compound activity. Efficacy was confirmed in immunotherapy-resistant organoids and patient-derived xenograft (PDX) models. The discovery of potent, selective covalent WRN inhibitors provides proof of concept for synthetic-lethal targeting of WRN in MSI cancer and tools to dissect WRN biology. Citation Format: Gabriele Picco, Yanhua Rao, Angham Al Saedi, Yang Lee, Shriram Bhosle, Carmen Herranz-Ors, Samantha Walker, Kieron May, Matthew Coelho, Jon Houseley, Benjamin Schwartz, Mathew J. Garnett. Novel WRN helicase inhibitors selectively target microsatellite unstable cancer cells [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Expanding and Translating Cancer Synthetic Vulnerabilities; 2024 Jun 10-13; Montreal, Quebec, Canada. Philadelphia (PA): AACR; Mol Cancer Ther 2024;23(6 Suppl):Abstract nr B020.

Searching Questions and Learning Problems in Large Problem Banks: Constructing Tests and Assignments on the Fly

Modern advances in creating shared banks of learning problems and automatic question and problem generation have led to the creation of large question banks in which human teachers cannot view every question. These questions are classified according to the knowledge necessary to solve them and the question difficulties. Constructing tests and assignments on the fly at the teacher’s request eliminates the possibility of cheating by sharing solutions because each student receives a unique set of questions. However, the random generation of predictable and effective assignments from a set of problems is a non-trivial task. In this article, an algorithm for generating assignments based on teachers’ requests for their content is proposed. The algorithm is evaluated on a bank of expression-evaluation questions containing more than 5000 questions. The evaluation shows that the proposed algorithm can guarantee the minimum expected number of target concepts (rules) in an exercise with any settings. The available bank and exercise difficulty chiefly determine the difficulty of the found questions. It almost does not depend on the number of target concepts per item in the exercise: teaching more rules is achieved by rotating them among the exercise items on lower difficulty settings. An ablation study show that all the principal components of the algorithm contribute to its performance. The proposed algorithm can be used to reliably generate individual exercises from large, automatically generated question banks according to teachers’ requests, which is important in massive open online courses.

Hohe und weiter steigende Ungleichheit in den USA – politisch nicht gewollt und doch geduldet?

Abstract Looking at both the Gini coefficients ex-ante as well as at the Gini coefficients ex-post reveals that the distribution of personal incomes has become more unequal in the USA in the periods between 1947 and 2020. This paper investigates the possible underlying concepts of income distribution targets and argues, based on empirical findings, that neither clear electoral/partisan cycles nor station-arity (equilibrium) in the figures of income distribution of the USA can be detected. Instead, one observes an unbroken cyclical upward trend in inequality, which contributes significantly to the fission in the US society and economy. Surprisingly, as measured by the results produced by US presidents on the national level since 1947, the Republican Party and the Democratic Party have made similar contributions to lowering the inequality.

Implementing a natural language processing pipeline to expedite clinical chart review to assess germline genetic testing in veterans with breast cancer.

e13640 Background: The Making Advances in Mammography and Medical Options for Veterans Act of 2022 mandated that the Department of Veterans Affairs (VA) perform a study of germline genetic testing in Veterans with breast cancer (BC). Assessing whether a Veteran with BC was offered germline genetic testing based on electronic health records requires time-consuming review of unstructured clinical notes by clinical experts who are also extremely busy providing care. To expedite this review process, we implemented a productionized natural language processing (NLP) pipeline to identify relevant concepts and highlight them within a custom-made chart review interface. Methods: Clinical subject-matter experts (SMEs) were convened to identify keywords suggestive of genetic testing. These were categorized into 3 sets: gene-related concepts, VA-specific services, and exams and tests. We then used entity recognition and entity linking to link these concepts to a knowledge base. These were enriched with entities similarly extracted from an open-source reference clinical text on germline genetic testing in BC. The resulting concepts were visualized as a word cloud and reviewed by the SMEs. We implemented a Power BI and Power Apps interface to display clinical notes from 200 randomly selected Veterans, with relevant concepts highlighted. Each Veteran’s notes were reviewed independently by 2 of the 5 SMEs, who assessed whether the Veteran was offered testing, referred to a genetic counseling service, or otherwise had evidence of testing. Disagreements were adjudicated by the whole team. Results: 11622 notes (mean 58/Veteran, 3028 characters/note) were reviewed. Table shows the most prevalent entities among genes, mutations, and exam names. Of the 5 reviewers, 3 annotated 80 patients each in a single session, lasting less than 50 minutes on average. Reviewers had an inter-annotator agreement of 85.5%. Mention of target concepts was associated with germline genetic testing with an accuracy of 80.4%, precision of 76.4%, recall of 99.1%, and F1 of 86.3%. Germline genetic testing rates are detailed in a separate submission. Conclusions: Modern informatics tools such as productionized NLP, the Microsoft Power Platform, and high-performance computing environments allow for rapid implementation of health information technology tools that can enable clinicians to gain critical insights quickly and efficiently from electronic health record data. [Table: see text]

DecomCAM: Advancing beyond saliency maps through decomposition and integration

Interpreting complex deep networks, notably pre-trained vision-language models (VLMs), is a formidable challenge. Current Class Activation Map (CAM) methods highlight regions revealing the model’s decision-making basis but lack clear saliency maps and detailed interpretability. To bridge this gap, we propose DecomCAM, a novel decomposition-and-integration method that distills shared patterns from channel activation maps. Utilizing singular value decomposition, DecomCAM decomposes class-discriminative activation maps into orthogonal sub-saliency maps (OSSMs), which are then integrated together based on their contribution to the target concept. Extensive experiments on six benchmarks reveal that DecomCAM not only excels in locating accuracy but also achieves an optimizing balance between interpretability and computational efficiency. Further analysis unveils that OSSMs correlate with discernible object components, facilitating a granular understanding of the model’s reasoning. This positions DecomCAM as a potential tool for fine-grained interpretation of advanced deep learning models. The code is avaible at https://github.com/CapricornGuang/DecomCAM.

A high-intensity, low-energy heavy ion source for a neutron target proof-of-principle experiment at LANSCE

The capability to directly study neutron capture reactions on radionuclides with half-lives on the of order minutes would allow key cross section measurements in nuclear astrophysics and energy applications. However, such experiments with stationary targets are currently impossible because of signal detection and target sample fielding issues. To overcome these challenges, a neutron target facility is being developed to permit neutron capture experiments on unstable isotopes in inverse kinematics at the Los Alamos Neutron Science Center (LANSCE). This next-generation facility will consist of a heavily moderated, high-intensity spallation neutron target coupled with a radioactive ion beam storage ring. A proof-of-principle experiment is underway to demonstrate this neutron target concept at LANSCE in the near term with stable ions and without a storage ring. Here, mA-level beams of 10-50 keV heavy ions exhibiting large resonant neutron capture cross sections will be generated by a new ion source, transported through a large-volume neutron moderator surrounding an adjacent spallation target driven by the LANSCE accelerator, and collected downstream of the moderator for subsequent decay-counting measurements. The neutron density within the moderator will be obtained from these decay yields and compared with Monte Carlo N-Particle (MCNP) simulation results. The science application, operational requirements, and performance objectives for this heavy ion source will be presented.

Exploring Attitudes toward Sustainability Education in a Group of Italian Preservice Teachers: The Role of Environmental Identity and Sense of Community Responsibility

In the current educational context, international institutions are being called to rethink education and declining new strategies that aim to teach sustainable behaviors. From this perspective, the present study aims to investigate the attitudes towards the sustainability of trainee teachers, verifying their connection with psycho-social variables, such as environmental identity and sense of community responsibility, and exploring their most common dispositions. In total, 126 Italian preservice teachers were involved in a mixed method study where they were asked to complete a questionnaire and participated in focus groups. The use of a combination of quantitative–qualitative methodologies has allowed us to obtain in-depth information about sustainability education. Quantitative data showed a virtuous circular relationship between environmental identity, attitudes towards education for sustainability, and the sense of community responsibility. Qualitative data showed the high frequency of two key action verbs, avoid and respect, but no specific reference to the educational field. After creating the target concept ‘to teach’, we observed a worrying absence of lemmas related to emotional dimension, and that the communication perspective is set on adults’ point of view, rather than on pupils. Limitations, strengths, and practical implications have been extensively discussed, especially in relation to the field of teacher education.

NKX2-5 regulates vessel remodeling in scleroderma-associated pulmonary arterial hypertension.

NKX2-5 is a member of the homeobox-containing transcription factors critical in regulating tissue differentiation in development. Here, we report a role for NKX2-5 in vascular smooth muscle cell phenotypic modulation in vitro and in vascular remodeling in vivo. NKX2-5 is upregulated in scleroderma patients with pulmonary arterial hypertension. Suppression of NKX2-5 expression in smooth muscle cells halted vascular smooth muscle proliferation and migration, enhanced contractility, and blocked the expression of extracellular matrix genes. Conversely, overexpression of NKX2-5 suppressed the expression of contractile genes (ACTA2, TAGLN, CNN1) and enhanced the expression of matrix genes (COL1) in vascular smooth muscle cells. In vivo, conditional deletion of NKX2-5 attenuated blood vessel remodeling and halted the progression to hypertension in a mouse chronic hypoxia model. This study revealed that signals related to injury such as serum and low confluence, which induce NKX2-5 expression in cultured cells, is potentiated by TGF-β and further enhanced by hypoxia. The effect of TGF-β was sensitive to ERK5 and PI3K inhibition. Our data suggest a pivotal role for NKX2-5 in the phenotypic modulation of smooth muscle cells during pathological vascular remodeling and provide proof of concept for therapeutic targeting of NKX2-5 in vasculopathies.

Glutamate Receptors and C-ABL Inhibitors: A New Therapeutic Approach for Parkinson's Disease.

Parkinson's disease (PD) is the second-most prevalent central nervous system (CNS) neurodegenerative condition. Over the past few decades, suppression of BCR-Abelson tyrosine kinase (c-Abl), which serves as a marker of -synuclein aggregation and oxidative stress, has shown promise as a potential therapy target in PD. c-Abl inhibition has the potential to provide neuroprotection against PD, as shown by experimental results and the first-in-human trial, which supports the strategy in bigger clinical trials. Furthermore, glutamate receptors have also been proposed as potential therapeutic targets for the treatment of PD since they facilitate and regulate synaptic neurotransmission throughout the basal ganglia motor system. It has been noticed that pharmacological manipulation of the receptors can change normal as well as abnormal neurotransmission in the Parkinsonian brain. The review study contributes to a comprehensive understanding of the approach toward the role of c-Abl and glutamate receptors in Parkinson's disease by highlighting the significance and urgent necessity to investigate new pharmacotherapeutic targets. The article covers an extensive insight into the concept of targeting, pathophysiology, and c-Abl interaction with α-synuclein, parkin, and cyclin-dependent kinase 5 (Cdk5). Furthermore, the concepts of Nmethyl- D-aspartate (NMDA), α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid receptor (AMPA) receptor, and glutamate receptors are discussed briefly. Conclusion: This review article focuses on in-depth literature findings supported by an evidence-based discussion on pre-clinical trials and clinical trials related to c-Abl and glutamate receptors that act as potential therapeutic targets for PD.

Fine-Grained Interpretability for EEG Emotion Recognition: Concat-Aided Grad-CAM and Systematic Brain Functional Network

EEG emotion recognition plays a significant role in various mental health services. Deep learning-based methods perform excellently, but still suffer from interpretability. Although methods such as Gradient-weighted Class Activation Mapping(Grad-CAM) can cope with the above problem, their coarse granularity cannot accurately reveal the mechanism to promote emotional intelligence. In this paper, fine-grained interpretability is proposed, called Concat-aided Grad-CAM. Specifically, the multi-level feature mapping before the fully connected layer is concatenated to obtain the gradients of the target concept so that the discriminant information can be directly located in the high-precision area. Unlike coarse-grained interpretability methods applied in EEG emotion recognition, it can accurately highlight the EEG channels related to emotion rather than an obscure area. In addition, a systematic brain functional network is proposed to reveal the relationship between those channels and to further improve emotion recognition performance. The channels with greater contributions are connected, and those connections are learned by dynamic graph convolutional networks, while the others are independent to eliminate interference. Experiments on two EEG emotion recognition datasets manifest that Concat-aided Grad-CAM can be interpreted by the fine-grained. In addition, it has been shown that the learned brain functional network can improve the performance of the baselines. Significantly, the experiment results achieve state-of-the-art performance in subject-dependent experiments.

All but One: Surgical Concept Erasing with Model Preservation in Text-to-Image Diffusion Models

Text-to-Image models such as Stable Diffusion have shown impressive image generation synthesis, thanks to the utilization of large-scale datasets. However, these datasets may contain sexually explicit, copyrighted, or undesirable content, which allows the model to directly generate them. Given that retraining these large models on individual concept deletion requests is infeasible, fine-tuning algorithms have been developed to tackle concept erasing in diffusion models. While these algorithms yield good concept erasure, they all present one of the following issues: 1) the corrupted feature space yields synthesis of disintegrated objects, 2) the initially synthesized content undergoes a divergence in both spatial structure and semantics in the generated images, and 3) sub-optimal training updates heighten the model's susceptibility to utility harm. These issues severely degrade the original utility of generative models. In this work, we present a new approach that solves all of these challenges. We take inspiration from the concept of classifier guidance and propose a surgical update on the classifier guidance term while constraining the drift of the unconditional score term. Furthermore, our algorithm empowers the user to select an alternative to the erasing concept, allowing for more controllability. Our experimental results show that our algorithm not only erases the target concept effectively but also preserves the model’s generation capability.