Multi-level Alignment Research Articles

Digging deeper: Unpacking the subnational political drivers of Chinese extractive investment in Latin America

There is a long scholarly tradition that argues that political variables figure prominently in international corporations’ decisions to invest abroad. Most of this literature has focused on the national political determinants of investment allocation. Yet, in some sectors, the bulk of this investment is local. Chief among these sectors are extractive industries where assets are endowed to different local jurisdictions. How do political factors shape extractive foreign direct investment (EFDI) in local jurisdictions? We focus on Latin America, a region of the world that is rich in natural resources, and that has recently attracted a sizable amount of Chinese investment. Drawing on a novel dataset on subnational Chinese EFDI in Argentina, Brazil, and Mexico (2002–2018), and using a logistic model with time series cross-national data we find that multilevel partisan alignment is a strong determinant of Chinese EFDI. In highly decentralized countries, Chinese EFDI is primarily targeted to subnational districts ruled by governors who are politically aligned with presidents. We illustrate the logics of this allocation with two case studies. Our findings make a valuable contribution to three growing fields of study: the subnational determinants of foreign investments, multilevel governance, and the political factors driving Chinese investments in developing countries.

Multi-level domain perturbation for source-free object detection in remote sensing images

ABSTRACT Recent advancements in cross-domain object detection have primarily relied on unsupervised domain adaptation (UDA) techniques to bridge domain gaps in remote sensing images. However, traditional UDA approaches require access to source domain data, creating privacy and transmission barriers. In response, we first introduce a novel setting called source-free object detection (SFOD) for remote sensing images. SFOD exclusively leverages pre-trained models from the source domain and unlabeled data from the target domain for adaptation. Our method emphasizes the detection model’s focus on domain-invariant features by ensuring consistency between the perturbed and target domains. We initiate domain perturbation by manipulating domain-specific features at both image and feature levels, guided by considerations of style and color biases. Subsequently, we employ a multi-level alignment mechanism to ensure label-level and feature-level consistency between the perturbed and target domains. Our method incorporates the Mean Teachers framework and prototype-based feature distillation to alleviate noise arising from pseudo-labels. Extensive experiments conducted across various adaptation scenarios validate the effectiveness of our approach.

Multi-level alignment processes in the sustainment of a youth substance use treatment model following a federal implementation initiative: A mixed method study

IntroductionGovernment agencies have identified evidence-based practice (EBP) dissemination as a pathway to high-quality behavioral health care for youth. However, gaps remain about how to best sustain EBPs in treatment organizations in the U.S., especially in resource-constrained settings like publicly-funded youth substance use services. One important, but understudied, determinant of EBP sustainment is alignment: the extent to which multi-level factors that influence sustainment processes and outcomes are congruent, consistent, and/or coordinated. This study examined the role of alignment in U.S. states' efforts to sustain the Adolescent Community Reinforcement Approach (A-CRA), an EBP for youth substance use disorders, during the COVID-19 pandemic. MethodsIn this mixed methods study, the qualitative investigation preceded and informed the quantitative investigation. We interviewed state administrators and providers (i.e., supervisors and clinicians) from 15 states that had completed a federal A-CRA implementation grant; providers also completed surveys. The sample included 50 providers from 35 treatment organizations that reported sustaining A-CRA when the COVID-19 pandemic began, and 20 state administrators. In qualitative thematic analyses, we applied the EPIS (Exploration, Preparation, Implementation, Sustainment) framework to characterize alignment processes that interviewees described as influential on sustainment. We then used survey items to quantitatively explore the associations described in qualitative themes, using bivariate linear regressions. ResultsAt the time of interview, staff from 80 % of the treatment organizations (n = 28), reported sustaining A-CRA. Providers from both sustainer and non-sustainer organizations, as well as state administrators, described major sources of misalignment when state agencies ceased technical assistance post-grant, and because limited staff capacity conflicted with A-CRA's training model, which was perceived as time-intensive. Participants described the pandemic as exacerbating preexisting challenges, including capacity issues. Sustainer organizations reported seeking new funding to help sustain A-CRA. Quantitative associations between self-rated extent of sustainment and other survey items mostly followed the pattern predicted from the qualitative findings. ConclusionsThe COVID-19 pandemic amplified longstanding A-CRA sustainment challenges, but treatment organizations already successfully sustaining A-CRA pre-pandemic largely continued. There are missed opportunities for state-level actors to coordinate with providers on the shared goal of EBP sustainment. A greater focus on alignment processes in research and practice could help states and providers strengthen sustainability planning.

Multi-Level Cross-Modal Alignment for Image Clustering

Recently, the cross-modal pretraining model has been employed to produce meaningful pseudo-labels to supervise the training of an image clustering model. However, numerous erroneous alignments in a cross-modal pretraining model could produce poor-quality pseudo labels and degrade clustering performance. To solve the aforementioned issue, we propose a novel Multi-level Cross-modal Alignment method to improve the alignments in a cross-modal pretraining model for downstream tasks, by building a smaller but better semantic space and aligning the images and texts in three levels, i.e., instance-level, prototype-level, and semantic-level. Theoretical results show that our proposed method converges, and suggests effective means to reduce the expected clustering risk of our method. Experimental results on five benchmark datasets clearly show the superiority of our new method.

Generalizable Sleep Staging via Multi-Level Domain Alignment

Automatic sleep staging is essential for sleep assessment and disorder diagnosis. Most existing methods depend on one specific dataset and are limited to be generalized to other unseen datasets, for which the training data and testing data are from the same dataset. In this paper, we introduce domain generalization into automatic sleep staging and propose the task of generalizable sleep staging which aims to improve the model generalization ability to unseen datasets. Inspired by existing domain generalization methods, we adopt the feature alignment idea and propose a framework called SleepDG to solve it. Considering both of local salient features and sequential features are important for sleep staging, we propose a Multi-level Feature Alignment combining epoch-level and sequence-level feature alignment to learn domain-invariant feature representations. Specifically, we design an Epoch-level Feature Alignment to align the feature distribution of each single sleep epoch among different domains, and a Sequence-level Feature Alignment to minimize the discrepancy of sequential features among different domains. SleepDG is validated on five public datasets, achieving the state-of-the-art performance.

Multi-level textual-visual alignment and fusion network for multimodal aspect-based sentiment analysis

Multimodal Aspect-Based Sentiment Analysis (MABSA) is an essential task in sentiment analysis that has garnered considerable attention in recent years. Typical approaches in MABSA often utilize cross-modal Transformers to capture interactions between textual and visual modalities. However, bridging the semantic gap between modalities spaces and addressing interference from irrelevant visual objects at different scales remains challenging. To tackle these limitations, we present the Multi-level Textual-Visual Alignment and Fusion Network (MTVAF) in this work, which incorporates three auxiliary tasks. Specifically, MTVAF first transforms multi-level image information into image descriptions, facial descriptions, and optical characters. These are then concatenated with the textual input to form a textual+visual input, facilitating comprehensive alignment between visual and textual modalities. Next, both inputs are fed into an integrated text model that incorporates relevant visual representations. Dynamic attention mechanisms are employed to generate visual prompts to control cross-modal fusion. Finally, we align the probability distributions of the textual input space and the textual+visual input space, effectively reducing noise introduced during the alignment process. Experimental results on two MABSA benchmark datasets demonstrate the effectiveness of the proposed MTVAF, showcasing its superior performance compared to state-of-the-art approaches. Our codes are available at https://github.com/MKMaS-GUET/MTVAF.

Demonstrating the deep institutionalisation of de facto responsible research and innovation (rri) in participatory market contexts: examples from Bolivia and the Netherlands

ABSTRACT Our paper reprises the concept deep institutionalisation of responsible innovation considering why and how it matters to add the adjective ‘deep’. We distinguish de facto responsible research and innovation (rri) as the study of how actors frame and govern responsibility through existing practices (‘in the wild’) and investigate how these practices become institutionalised. We present a diagnostic framework comprising four axes which facilitates the critical and reflexive empirical interrogation of deep institutionalisation (DI). Deploying the framework, the paper explores DI in two very different cases: an inter-organisational case of Participatory Guarantee Systems (PGS) in Bolivia; and an intra-organisational case of societal engagement within TNO in the Netherlands. Controversially perhaps, we argue that normative features of responsibility are enacted, amplified and potentially institutionalised through markets. Both cases show how particular market features become recursively qualified through the four mutually reinforcing processes that comprise deep institutionalisation: (i) historical contingency; (ii) institutional amplification; (iii) systemic overflowing; and (iv)multi-level alignment.

Multi-level multilingual semantic alignment for zero-shot cross-lingual transfer learning

Recently, cross-lingual transfer learning has attracted extensive attention from both academia and industry. Previous studies usually focus only on the single-level alignment (e.g., word-level, sentence-level), based on pre-trained language models. However, it leads to suboptimal performance in downstream tasks of the low-resource language due to the missing correlation of hierarchical semantic information (e.g., sentence-to-word, word-to-word). Therefore, in this paper, we propose a novel multi-level alignment framework, which hierarchically learns the semantic correlation between multiple levels by leveraging well-designed alignment training tasks. In addition, we devise an attention-based fusion mechanism (AFM) to infuse semantic information from high levels. Extensive experiments on mainstream cross-lingual tasks (e.g., text classification, paraphrase identification, and named entity recognition) demonstrate the effectiveness of our proposed method, and also show that our model achieves state-of-the-art performance across various benchmarks compared to other strong baselines.

Hierarchical matching and reasoning for multi-query image retrieval

As a promising field, Multi-Query Image Retrieval (MQIR) aims at searching for the semantically relevant image given multiple region-specific text queries. Existing works mainly focus on a single-level similarity between image regions and text queries, which neglect the hierarchical guidance of multi-level similarities and result in incomplete alignments. Besides, the high-level semantic correlations that intrinsically connect different region–query pairs are rarely considered. To address above limitations, we propose a novel Hierarchical Matching and Reasoning Network (HMRN) for MQIR. It disentangles MQIR into three hierarchical semantic representations, which is responsible to capture fine-grained local details, contextual global scopes, and high-level inherent correlations. HMRN consists of two modules: Scalar-based Matching (SM) module and Vector-based Reasoning (VR) module. Specifically, the SM module characterizes the multi-level alignment similarity, which consists of a fine-grained local-level similarity and a context-aware global-level similarity. Afterwards, the VR module is developed to excavate the potential semantic correlations among multiple region–query pairs, which further explores the high-level reasoning similarity. Finally, these three-level similarities are aggregated into a joint similarity space to form the ultimate similarity. Extensive experiments on the benchmark dataset demonstrate that our HMRN substantially surpasses the current state-of-the-art methods. For instance, compared with the existing best method Drill-down, the metric R@1 in the last round is improved by 23.4%. Our source codes will be released at https://github.com/LZH-053/HMRN.

Semantic enhancement and multi-level alignment network for cross-modal retrieval

Semantic enhancement and multi-level alignment network for cross-modal retrieval

MLFA: Towards Realistic Test Time Adaptive Object Detection by Multi-level Feature Alignment.

Object detection methods have achieved remarkable performances when the training and testing data satisfy the assumption of i.i.d. However, the training and testing data may be collected from different domains, and the gap between the domains can significantly degrade the detectors. Test Time Adaptive Object Detection (TTA-OD) is a novel online approach that aims to adapt detectors quickly and make predictions during the testing procedure. TTA-OD is more realistic than the existing unsupervised domain adaptation and source-free unsupervised domain adaptation approaches. For example, self-driving cars need to improve their perception of new environments in the TTA-OD paradigm during driving. To address this, we propose a multi-level feature alignment (MLFA) method for TTA-OD, which is able to adapt the model online based on the steaming target domain data. For a more straightforward adaptation, we select informative foreground and background features from image feature maps and capture their distributions using probabilistic models. Our approach includes: i) global-level feature alignment to align all informative feature distributions, thereby encouraging detectors to extract domain-invariant features, and ii) cluster-level feature alignment to match feature distributions for each category cluster across different domains. Through the multi-level alignment, we can prompt detectors to extract domain-invariant features, as well as align the category-specific components of image features from distinct domains. We conduct extensive experiments to verify the effectiveness of our proposed method. Our code is accessible at https://github.com/yaboliudotug/MLFA.

Multi-Level Linguistic Alignment in a Dynamic Collaborative Problem-Solving Task.

Cocreating meaning in collaboration is challenging. Success is often determined by people's abilities to coordinate their language to converge upon shared mental representations. Here we explore one set of low-level linguistic behaviors, linguistic alignment, that both emerges from, and facilitates, outcomes of high-level convergence. Linguistic alignment captures the ways people reuse, that is, "align to," the lexical, syntactic, and semantic forms of others' utterances. Our focus is on the temporal change of multi-level linguistic alignment, as well as how alignment is related to communicative outcomes within a unique collaborative problem-solving paradigm. The primary task, situated within a virtual educational video game, requires creative thinking between three people where the paths for possible solutions are highly variable. We find that over time interactions are marked by decreasing lexical and syntactic alignment, with a trade-off of increasing semantic alignment. However, greater semantic alignment does not translate into better team performance. Overall, these findings provide greater clarity on the role of linguistic coordination within complex and dynamic collaborative problem-solvingtasks.

A Multi-Level Alignment and Cross-Modal Unified Semantic Graph Refinement Network for Conversational Emotion Recognition

A Multi-Level Alignment and Cross-Modal Unified Semantic Graph Refinement Network for Conversational Emotion Recognition

Multi-Level Objective Alignment Transformer for Fine-Grained Oral Panoramic X-Ray Report Generation

Multi-Level Objective Alignment Transformer for Fine-Grained Oral Panoramic X-Ray Report Generation

From optical to SAR： A SAR ship detection algorithm based on multi-level cross-modality alignment

From optical to SAR： A SAR ship detection algorithm based on multi-level cross-modality alignment

Image-Specific Information Suppression and Implicit Local Alignment for Text-Based Person Search.

Text-based person search (TBPS) is a challenging task that aims to search pedestrian images with the same identity from an image gallery given a query text. In recent years, TBPS has made remarkable progress, and state-of-the-art (SOTA) methods achieve superior performance by learning local fine-grained correspondence between images and texts. However, most existing methods rely on explicitly generated local parts to model fine-grained correspondence between modalities, which is unreliable due to the lack of contextual information or the potential introduction of noise. Moreover, the existing methods seldom consider the information inequality problem between modalities caused by image-specific information. To address these limitations, we propose an efficient joint multilevel alignment network (MANet) for TBPS, which can learn aligned image/text feature representations between modalities at multiple levels, and realize fast and effective person search. Specifically, we first design an image-specific information suppression (ISS) module, which suppresses image background and environmental factors by relation-guided localization (RGL) and channel attention filtration (CAF), respectively. This module effectively alleviates the information inequality problem and realizes the alignment of information volume between images and texts. Second, we propose an implicit local alignment (ILA) module to adaptively aggregate all pixel/word features of image/text to a set of modality-shared semantic topic centers and implicitly learn the local fine-grained correspondence between modalities without additional supervision and cross-modal interactions. Also, a global alignment (GA) is introduced as a supplement to the local perspective. The cooperation of global and local alignment modules enables better semantic alignment between modalities. Extensive experiments on multiple databases demonstrate the effectiveness and superiority of our MANet.

Multi-Level Alignments for Compressed Video Super-Resolution

Multi-Level Alignments for Compressed Video Super-Resolution

Rethinking automatic segmentation of gross target volume from a decoupling perspective

Accurate and reliable segmentation of Gross Target Volume (GTV) is critical in cancer Radiation Therapy (RT) planning, but manual delineation is time-consuming and subject to inter-observer variations. Recently, deep learning methods have achieved remarkable success in medical image segmentation. However, due to the low image contrast and extreme pixel imbalance between GTV and adjacent tissues, most existing methods usually obtained limited performance on automatic GTV segmentation. In this paper, we propose a Heterogeneous Cascade Framework (HCF) from a decoupling perspective, which decomposes the GTV segmentation into independent recognition and segmentation subtasks. The former aims to screen out the abnormal slices containing GTV, while the latter performs pixel-wise segmentation of these slices. With the decoupled two-stage framework, we can efficiently filter normal slices to reduce false positives. To further improve the segmentation performance, we design a multi-level Spatial Alignment Network (SANet) based on the feature pyramid structure, which introduces a spatial alignment module into the decoder to compensate for the information loss caused by downsampling. Moreover, we propose a Combined Regularization (CR) loss and Balance-Sampling Strategy (BSS) to alleviate the pixel imbalance problem and improve network convergence. Extensive experiments on two public datasets of StructSeg2019 challenge demonstrate that our method outperforms state-of-the-art methods, especially with significant advantages in reducing false positives and accurately segmenting small objects. The code is available at https://github.com/shijun18/GTV_AutoSeg.

AGE-FRIENDLY CITIES AND COMMUNITIES: RESEARCH ON LOCAL PRACTICE TO STRENGTHEN THE MOVEMENT’S FUTURE

Abstract The age-friendly cities and communities (AFCC) movement has accelerated a major paradigmatic shift for the field of aging in the 21st century. Launched and spurred by high-profile government agencies and organizations across the globe, this ambitious movement highlights the transformative potential of deliberate, cross-sectoral efforts to improve built, social, and service environments of localities in response to population aging. This presentation will provide a brief overview of the development of a global program of policy and practice on AFCCs, with an emphasis on its emergence in the United States. It will then present insights regarding challenges and opportunities for the future based on findings from an ongoing, longitudinal study of a regional network of AFCC initiatives. The study, designed as a developmental evaluation, explores the perspectives of on-the-ground AFCC leaders on their practice as part of a private philanthropic grantmaking initiative to spur AFCCs in New Jersey. Reflecting on our years of engaged research with this network, the presentation will highlight issues concerning multi-level systems alignment, the promises and pitfalls of public sector leadership, the need for additional practice theories, and greater incorporation of considerations of aging equity and racial justice. The presentation will demonstrate the importance of continuing to leverage the strengths of community-engaged research to better ensure the movement’s relevance, reach, and equitable impact across diverse community, organizational, and broader systems contexts.

Multi-level alignment for few-shot temporal action localization

Multi-level alignment for few-shot temporal action localization