Use of Interactive Food Labels to Increase Confidence in Preparing Produce among College Students.

College students may have limited access to produce and may lack confidence in preparing it, but cooking videos can show how to make healthy dishes. The Cognitive Theory of Multimedia Learning suggests that learning is enhanced when visual and auditory information is presented considering cognitive load (e.g., highlighting important concepts, eliminating extraneous information, and keeping the video brief and conversational). The purpose of this project was to pilot test a food label for produce grown at an urban university and assess whether student confidence in preparing produce improved after using the label and QR code to view a recipe video developed using principles from the Cognitive Theory of Multimedia Learning. The video showed a student preparing a salad with ingredients available on campus. Students indicated the label was helpful and reported greater perceived confidence in preparing lettuce after viewing the label and video (mean confidence of 5.60 ± 1.40 before vs. 6.14 ± 0.89 after, p = 0.016, n = 28). Keeping the video short and providing ingredients and amounts onscreen as text were cited as helpful. Thus, a brief cooking video and interactive label may improve confidence in preparing produce available on campus. Future work should determine whether the label impacts produce consumption and if it varies depending on the type of produce used.

Adaptive ambiguity-aware weighting for multi-label recognition with limited annotations.

In multi-label recognition, effectively addressing the challenge of partial labels is crucial for reducing annotation costs and enhancing model generalization. Existing methods exhibit limitations by relying on unrealistic simulations with uniformly dropped labels, overlooking how ambiguous instances and instance-level factors impacts label ambiguity in real-world datasets. To address this deficiency, our paper introduces a realistic partial label setting grounded in instance ambiguity, complemented by Reliable Ambiguity-Aware Instance Weighting (R-AAIW)-a strategy that utilizes importance weighting to adapt dynamically to the inherent ambiguity of multi-label instances. The strategy leverages an ambiguity score to prioritize learning from clearer instances. As proficiency of the model improves, the weights are dynamically modulated to gradually shift focus towards more ambiguous instances. By employing an adaptive re-weighting method that adjusts to the complexity of each instance, our approach not only enhances the model's capability to detect subtle variations among labels but also ensures comprehensive learning without excluding difficult instances. Extensive experimentation across various benchmarks highlights our approach's superiority over existing methods, showcasing its ability to provide a more accurate and adaptable framework for multi-label recognition tasks.

Accessible Point-and-Tap Interaction for Acquiring Detailed Information about Tactile Graphics and 3D Models.

We have devised a novel "Point-and-Tap" interface that enables people who are blind or visually impaired (BVI) to easily acquire multiple levels of information about tactile graphics and 3D models. The interface uses an iPhone's depth and color cameras to track the user's hands while they interact with a model. When the user points to a feature of interest on the model with their index finger, the system reads aloud basic information about that feature. For additional information, the user lifts their index finger and taps the feature again. This process can be repeated multiple times to access additional levels of information. For instance, tapping once on a region in a tactile map could trigger the name of the region, with subsequent taps eliciting the population, area, climate, etc. No audio labels are triggered unless the user makes a pointing gesture, which allows the user to explore the model freely with one or both hands. Multiple taps can be used to skip through information levels quickly, with each tap interrupting the current utterance. This allows users to reach the desired level of information more quickly than listening to all levels in sequence. Experiments with six BVI participants demonstrate that the approach is practical, easy to learn and effective.

Harnessing large language models' zero-shot and few-shot learning capabilities for regulatory research.

Large language models (LLMs) are sophisticated AI-driven models trained on vast sources of natural language data. They are adept at generating responses that closely mimic human conversational patterns. One of the most notable examples is OpenAI's ChatGPT, which has been extensively used across diverse sectors. Despite their flexibility, a significant challenge arises as most users must transmit their data to the servers of companies operating these models. Utilizing ChatGPT or similar models online may inadvertently expose sensitive information to the risk of data breaches. Therefore, implementing LLMs that are open source and smaller in scale within a secure local network becomes a crucial step for organizations where ensuring data privacy and protection has the highest priority, such as regulatory agencies. As a feasibility evaluation, we implemented a series of open-source LLMs within a regulatory agency's local network and assessed their performance on specific tasks involving extracting relevant clinical pharmacology information from regulatory drug labels. Our research shows that some models work well in the context of few- or zero-shot learning, achieving performance comparable, or even better than, neural network models that needed thousands of training samples. One of the models was selected to address a real-world issue of finding intrinsic factors that affect drugs' clinical exposure without any training or fine-tuning. In a dataset of over 700 000 sentences, the model showed a 78.5% accuracy rate. Our work pointed to the possibility of implementing open-source LLMs within a secure local network and using these models to perform various natural language processing tasks when large numbers of training examples are unavailable.

Mind Melds: Verbal Labels Induce Greater Representational Alignment.

What determines whether two people represent something in a similar way? We examined the role of verbal labels in promoting representational alignment. Across two experiments, three groups of participants sorted novel shapes from two visually dissimilar categories. Prior to sorting, participants in two of the groups were pre-exposed to the shapes using a simple visual matching task designed to reinforce the visual category structure. In one of these groups, participants additionally heard one of two nonsense category labels accompanying the shapes. Exposure to these redundant labels led people to represent the shapes in a more categorical way, which led to greater alignment between sorters. We found this effect of label-induced alignment despite the two categories being highly visually distinct and despite participants in both pre-exposure conditions receiving identical visual experience with the shapes. Experiment 2 replicated this basic result using more even more stringent testing conditions. The results hint at the possibly extensive role that labels may play in aligning people's mental representations.

Seeing the Forest but Naming the Trees: An Object-Over-Place Bias in Learning Noun Labels.

Objects and places are foundational spatial domains represented in human symbolic expressions, like drawings, which show a prioritization of depicting small-scale object-shape information over the large-scale navigable place information in which objects are situated. Is there a similar object-over-place bias in language? Across six experiments, adults and 3- to 4-year-old children were asked either to extend a novel noun in a labeling phrase, to extend a novel noun in a prepositional phrase, or to simply match pictures. To dissociate specific object and place information from more general figure and ground information, participants either saw scenes with both place information (a room) and object information (a block in the room), or scenes with two kinds of object information that matched the figure-ground relations of the room and block by presenting an open container with a smaller block inside. While adults showed a specific object-over-place bias in both extending novel noun labels and matching, they did not show this bias in extending novel nouns following prepositions. Young children showed this bias in extending novel noun labels only. Spatial domains may thus confer specific and foundational biases for word learning that may change through development in a way that is similar to that of other word-learning biases about objects, like the shape bias. These results expand the symbolic scope of prior studies on object biases in drawing to object biases in language, and they expand the spatial domains of prior studies characterizing the language of objects and places.

Helpful or harmful? The effect of a diagnostic label and its later retraction on person impressions.

Diagnostic labels for mental health conditions can inadvertently reinforce harmful stereotypes and exacerbate stigma. If a diagnosis is incorrect and a label is wrongly applied, this may negatively impact person impressions even if the inaccurate label is later corrected. This registered report examined this issue. Participants (N = 560) read a vignette about a hospital patient who was either diagnosed with schizophrenia, diagnosed with major depressive disorder, or not diagnosed with a mental health condition. The diagnostic labels were later retracted strongly, retracted weakly, or not retracted. Participants completed several stigma measures (desire for social distance, perceived dangerousness, and unpredictability), plus several inferential-reasoning measures that tested their reliance on the diagnostic label. As predicted, each mental health diagnosis elicited stigma, and influenced inferential reasoning. This effect was stronger for the schizophrenia diagnosis compared to the major depressive disorder diagnosis. Importantly, the diagnostic label continued to influence person judgments after a clear retraction (strong or weak), highlighting the limitations of corrections in reducing reliance on person-related misinformation and mental health stigma.

Evidence-based food serving size labelling: Survey and laboratory analyses of consumer cooking spray usage.

Nutrition label serving sizes are determined primarily based on typical consumption when such data are available. However, such data are not available for certain foods such as spray cooking oil (cooking spray). Our study assessed cooking spray use by the United States (US) adults compared to the 0.25-s serving size used on US-sold cooking spray labels. Adults (n = 1041, aged 33 ± 16.7 years) completed a 13-question survey on cooking spray use and perceptions. In the survey, participants reported using cooking spray for 1.9 ± 0.9 s per use, and 42.3%-43.1% of participants reported being more likely to purchase products if they were labelled calorie- or fat-free. Next, 30 adults (aged 29.7 ± 11.0 years) completed a laboratory-based study which assessed cooking spray durations for seven cookware items. Spray times ranged from 1.0 ± 0.5 (smallest pan) to 2.5 ± 1.3 s (largest baking sheet), with 100% of sprays (210/210) exceeding the 0.25-s US serving size. Our results suggest that cooking spray serving size should be increased to 1 s to better reflect actual consumption, and this would have the added benefit of aligning better with cooking spray serving sizes in other developed countries (0.5-1.0 s). A 1-s serving size would also preclude cooking spray advertised as calorie- or fat-free, allowing consumers to make more informed choices on the dietary implications of using cooking spray.

Effectiveness of alcohol warning labels for at-risk groups and the general public: A policy-informing randomized experiment in Chile.

OBJECTIVE: The World Health Organization recommends using health-risk warnings on alcoholic beverages. This study examines the impact of separate or combined warning labels for at-risk groups and the general population on alcohol purchase decisions. METHODS: In 2022, 7758 adults who consumed alcohol or were pregnant/lactating women (54.0 % female, mean age = 40.6 years) were presented with an online store's beverage section and randomly assigned to one of six warning labels in a between-subjects experimental design: no-warning, pregnant/lactating, drinking-driving, general cancer risk, combined warnings, and assorted warnings across bottles. The main outcome, the intention to purchase an alcoholic vs. non-alcoholic beverage, was examined with adjusted risk differences using logistic regressions. RESULTS: Participants exposed to the general cancer risk warning decreased their alcoholic choices by 10.4 percentage points (pp.) (95 % CI [-0.139, -0.069], p < 0.001, OR = 0.561), while those in the pregnancy/lactation warning condition did it by 3.8 pp. (95 % CI [-0.071, -0.005], p = 0.025, OR = 0.806). The driving-drinking warning had no significant effect. Participants exposed to the combined warnings label, or the assorted warnings reduced alcohol purchase decisions by 6.1 pp. (95 % CI [-0.095, -0.028], p < 0.001, OR = 0.708) and 4.3 pp. (95 % CI [-0.076, -0.010], p = 0.011, OR = 0.782), respectively. Cancer warning outperformed other labels and was effective for subgroups such as pregnant/lactating women, young adults, and low-income individuals. CONCLUSIONS: General cancer risk warnings are more effective at reducing alcohol purchase decisions compared to warning labels for specific groups or labels using multiple warnings. In addition to warning labels, other policies should be considered for addressing well-known alcohol-related risks (e.g., drinking and driving).

Perception is reality: qualitative insights into how consumers perceive alcohol warning labels.

AIMS: This study explores perspectives of on-pack alcohol warning labels, and how they might influence alcohol purchase and/or consumption behavior to inform culturally appropriate label design for effective behavior change. METHODS: New Zealand participants ≥18 years, who reported having purchased and consumed alcoholic beverages in the last month were recruited via a market research panel and grouped into 10 focus groups (n = 53) by ethnicity (general population, Maori, and Pacific peoples), age group, and level of alcohol consumption. Participants were shown six potential alcohol health warning labels, with design informed by relevant literature, label framework, and stakeholder feedback. Interviews were transcribed and analyzed via qualitative (directed) content analysis. RESULTS: Effective alcohol labels should be prominent, featuring large red and/or black text with a red border, combining text with visuals, and words like "WARNING" in capitals. Labels should contrast with bottle color, be easily understood, and avoid excessive text and confusing imagery. Participants preferred specific health outcomes, such as heart disease and cancer, increasing message urgency and relevance. Anticipated behavior change included reduced drinking and increased awareness of harms, but some may attempt to mitigate warnings by covering or removing labels. Contextual factors, including consistent design and targeted labels for different beverages and populations, are crucial. There was a strong emphasis on collective health impacts, particularly among Maori and Pacific participants. CONCLUSIONS: Our findings indicate that implementing alcohol warning labels, combined with comprehensive strategies like retail and social marketing campaigns, could effectively inform and influence the behavior of New Zealand's varied drinkers.

G-T correcting: an improved training of image segmentation under noisy labels.

Data-driven medical image segmentation networks require expert annotations, which are hard to obtain. Non-expert annotations are often used instead, but these can be inaccurate (referred to as "noisy labels"), misleading the network's training and causing a decline in segmentation performance. In this study, we focus on improving the segmentation performance of neural networks when trained with noisy annotations. Specifically, we propose a two-stage framework named "G-T correcting," consisting of "G" stage for recognizing noisy labels and "T" stage for correcting noisy labels. In the "G" stage, a positive feedback method is proposed to automatically recognize noisy samples, using a Gaussian mixed model to classify clean and noisy labels through the per-sample loss histogram. In the "T" stage, a confident correcting strategy and early learning strategy are adopted to allow the segmentation network to receive productive guidance from noisy labels. Experiments on simulated and real-world noisy labels show that this method can achieve over 90% accuracy in recognizing noisy labels, and improve the network's DICE coefficient to 91%. The results demonstrate that the proposed method can enhance the segmentation performance of the network when trained with noisy labels, indicating good clinical application prospects.

Introducing qrlabelr: Fast user-friendly software for machine- and human-readable labels in agricultural research and development.

The advent of modern tools in agricultural experiments, digital data collection, and high-throughput phenotyping have necessitated field plot labels that are both machine- and human-readable. Such labels are usually made with commercial software, which are often inaccessible to under-funded research programs in developing countries. The availability of free fit-for-purpose label design software to under-funded research programs in developing countries would address one of the main roadblocks to modernizing agricultural research. The goal was to develop a new open-source software with design features well-suited for field trials and other agricultural experiments. We report here qrlabelr, a new software for creating print-ready plot labels that builds on the foundation of an existing open-source program. The qrlabelr software offers more flexibility in the label design steps, guarantees true string fidelity after QR encoding, and provides faster label generation to users. The new software is available as an R package and offers customizable functions for generating plot labels. For non-R users or beginners in R programming, the package provides an interactive Shiny app version that can be launched from R locally or accessed online at https://bit.ly/3Sud4xy. The design philosophy of this new program emphasizes the adoption of best practices in plot label design to enhance reproducibility, tracking, and accurate data curation in agricultural research and development studies.

A semi-supervised segmentation method for microscopic hyperspectral pathological images based on multi-consistency learning.

Pathological images are considered the gold standard for clinical diagnosis and cancer grading. Automatic segmentation of pathological images is a fundamental and crucial step in constructing powerful computer-aided diagnostic systems. Medical microscopic hyperspectral pathological images can provide additional spectral information, further distinguishing different chemical components of biological tissues, offering new insights for accurate segmentation of pathological images. However, hyperspectral pathological images have higher resolution and larger area, and their annotation requires more time and clinical experience. The lack of precise annotations limits the progress of research in pathological image segmentation. In this paper, we propose a novel semi-supervised segmentation method for microscopic hyperspectral pathological images based on multi-consistency learning (MCL-Net), which combines consistency regularization methods with pseudo-labeling techniques. The MCL-Net architecture employs a shared encoder and multiple independent decoders. We introduce a Soft-Hard pseudo-label generation strategy in MCL-Net to generate pseudo-labels that are closer to real labels for pathological images. Furthermore, we propose a multi-consistency learning strategy, treating pseudo-labels generated by the Soft-Hard process as real labels, by promoting consistency between predictions of different decoders, enabling the model to learn more sample features. Extensive experiments in this paper demonstrate the effectiveness of the proposed method, providing new insights for the segmentation of microscopic hyperspectral tissue pathology images.

BPT-PLR: A Balanced Partitioning and Training Framework with Pseudo-Label Relaxed Contrastive Loss for Noisy Label Learning.

While collecting training data, even with the manual verification of experts from crowdsourcing platforms, eliminating incorrect annotations (noisy labels) completely is difficult and expensive. In dealing with datasets that contain noisy labels, over-parameterized deep neural networks (DNNs) tend to overfit, leading to poor generalization and classification performance. As a result, noisy label learning (NLL) has received significant attention in recent years. Existing research shows that although DNNs eventually fit all training data, they first prioritize fitting clean samples, then gradually overfit to noisy samples. Mainstream methods utilize this characteristic to divide training data but face two issues: class imbalance in the segmented data subsets and the optimization conflict between unsupervised contrastive representation learning and supervised learning. To address these issues, we propose a Balanced Partitioning and Training framework with Pseudo-Label Relaxed contrastive loss called BPT-PLR, which includes two crucial processes: a balanced partitioning process with a two-dimensional Gaussian mixture model (BP-GMM) and a semi-supervised oversampling training process with a pseudo-label relaxed contrastive loss (SSO-PLR). The former utilizes both semantic feature information and model prediction results to identify noisy labels, introducing a balancing strategy to maintain class balance in the divided subsets as much as possible. The latter adopts the latest pseudo-label relaxed contrastive loss to replace unsupervised contrastive loss, reducing optimization conflicts between semi-supervised and unsupervised contrastive losses to improve performance. We validate the effectiveness of BPT-PLR on four benchmark datasets in the NLL field: CIFAR-10/100, Animal-10N, and Clothing1M. Extensive experiments comparing with state-of-the-art methods demonstrate that BPT-PLR can achieve optimal or near-optimal performance.

Effectiveness of labels in digital art experience: psychophysiological and behavioral evidence.

Introduction: Nowadays museums make large use of digital materials (e.g., virtual tours) to attract visitors. Therefore, it is worthwhile investigating which variables affect the engagement with art outside the museum, and whether digital reproductions of artworks are as effective as museum originals in producing a satisfying aesthetic experience. Methods: Here we tested the effectiveness of introducing additional informative materials on the artistic enjoyment of contemporary paintings presented on a computer screen. Naïve observers were exposed to essential and descriptive labels before viewing artworks. We flanked traditional measurement methods - viewing times and questionnaires, with biometric parameters - pupil responses, eye movements, heart rate, and electrodermal activity. The results were then compared to our previous museum study that adopted the same experimental paradigm. Results: Our behavioral and psychophysiological data lead to a complex pattern of results. As found in the museum setting, providing detailed descriptions decreases complexity, evokes more positive sensations, and induces pupil dilation but does not enhance aesthetic appreciation. These results suggested that informative labels improve understanding and emotions but have a limited impact on the hedonic evaluation of artworks in both contexts. However, other results do not mirror those found in the museum; in the laboratory setting, participants spend a similar amount of time, have a comparable gaze behavior, and their electrodermal activity and heart rate do not change when viewing artworks with different types of labels. The main difference between the lab and museum settings is the shorter time spent viewing digital reproductions vs. real paintings, although subjective ratings (e.g., liking, interest) are comparable. Discussion: Overall, this study indicates that the environmental context does impact the aesthetic experience; although, some beneficial effects of introducing additional relevant content in labels accompanying artworks can also be acquainted through digital media outside of the museum.

Gamified Crowdsourcing as a Novel Approach to Lung Ultrasound Data Set Labeling: Prospective Analysis.

BACKGROUND: Machine learning (ML) models can yield faster and more accurate medical diagnoses; however, developing ML models is limited by a lack of high-quality labeled training data. Crowdsourced labeling is a potential solution but can be constrained by concerns about label quality. OBJECTIVE: This study aims to examine whether a gamified crowdsourcing platform with continuous performance assessment, user feedback, and performance-based incentives could produce expert-quality labels on medical imaging data. METHODS: In this diagnostic comparison study, 2384 lung ultrasound clips were retrospectively collected from 203 emergency department patients. A total of 6 lung ultrasound experts classified 393 of these clips as having no B-lines, one or more discrete B-lines, or confluent B-lines to create 2 sets of reference standard data sets (195 training clips and 198 test clips). Sets were respectively used to (1) train users on a gamified crowdsourcing platform and (2) compare the concordance of the resulting crowd labels to the concordance of individual experts to reference standards. Crowd opinions were sourced from DiagnosUs (Centaur Labs) iOS app users over 8 days, filtered based on past performance, aggregated using majority rule, and analyzed for label concordance compared with a hold-out test set of expert-labeled clips. The primary outcome was comparing the labeling concordance of collated crowd opinions to trained experts in classifying B-lines on lung ultrasound clips. RESULTS: Our clinical data set included patients with a mean age of 60.0 (SD 19.0) years; 105 (51.7%) patients were female and 114 (56.1%) patients were White. Over the 195 training clips, the expert-consensus label distribution was 114 (58%) no B-lines, 56 (29%) discrete B-lines, and 25 (13%) confluent B-lines. Over the 198 test clips, expert-consensus label distribution was 138 (70%) no B-lines, 36 (18%) discrete B-lines, and 24 (12%) confluent B-lines. In total, 99,238 opinions were collected from 426 unique users. On a test set of 198 clips, the mean labeling concordance of individual experts relative to the reference standard was 85.0% (SE 2.0), compared with 87.9% crowdsourced label concordance (P=.15). When individual experts' opinions were compared with reference standard labels created by majority vote excluding their own opinion, crowd concordance was higher than the mean concordance of individual experts to reference standards (87.4% vs 80.8%, SE 1.6 for expert concordance; P<.001). Clips with discrete B-lines had the most disagreement from both the crowd consensus and individual experts with the expert consensus. Using randomly sampled subsets of crowd opinions, 7 quality-filtered opinions were sufficient to achieve near the maximum crowd concordance. CONCLUSIONS: Crowdsourced labels for B-line classification on lung ultrasound clips via a gamified approach achieved expert-level accuracy. This suggests a strategic role for gamified crowdsourcing in efficiently generating labeled image data sets for training ML systems.

Can I still eat this? Using implicit and explicit measures to explore consumer behavior toward food products with date labels.

This study investigates implicit and explicit attitudes toward products before and beyond the best-before date (BBD) using an Implicit Association Test and an online questionnaire. Moreover, we test whether consumer perception of and behavior toward products beyond the BBD can be manipulated using a priming task. We use a three-group between-subjects design where respondents had to recall either a frugal, a wasteful, or an unrelated behavior. Results show that consumers have negative implicit associations with products beyond the BBD. Reduced health and safety perceptions, consumers' strategies to determine edibility, and general risk perception of products beyond the BBD predict consumption of these products. While recalling a frugal behavior does not have significant effects, recalling a wasteful behavior prior to evaluating products beyond the BBD leads to a decrease in the perceived safety and healthfulness of these products.

Self-reported decreases in the purchases of selected unhealthy foods resulting from the implementation of warning labels in Mexican youth and adult population.

BACKGROUND: Front-of-package nutritional warning labels (WLs) are designed to facilitate identification and selection of healthier food choices. We assessed self-reported changes in purchasing different types of unhealthy foods due to WLs in Mexico and the association between the self-reported reductions in purchases of sugary beverages and intake of water and sugar-sweetened beverages. METHODS: Data came from 14 to 17 year old youth (n = 1,696) and adults ≥ 18 (n = 7,775) who participated in the Mexican arm of the 2020-2021 International Food Policy Study, an annual repeat cross-sectional online survey. Participants self-reported whether the WLs had influenced them to purchase less of each of nine unhealthy food categories due to WLs. Among adults, a 23-item Beverage Frequency Questionnaire was used derive past 7-day intake of water and sugary beverages analyzed to determine the relationship between self-reported reductions in purchasing sugary drinks due to the WLs. Multilevel mixed-effects logistic regression models were fitted to estimate the percentage of participants who self-reported reducing purchases within each food group, and overall. Sociodemographic characteristics associated with this reduction were investigated as well. RESULTS: Overall, 44.8% of adults and 38.7% of youth reported buying less of unhealthy food categories due to the implementation of WL, with the largest proportion reporting decreased purchases of cola, regular and diet soda. A greater impact of WLs on the reported purchase of unhealthy foods was observed among the following socio-demographic characteristics: females, individuals who self-identified as indigenous, those who were overweight, individuals with lower educational levels, those with higher nutrition knowledge, households with children, and those with a significant role in household food purchases. In addition, adults who reported higher water intake and lower consumption of sugary beverages were more likely to report reduced purchases of sugary drinks due to the WLs. Adults who reported greater water intake and lower sugary beverages intake were significantly more likely to report buying fewer sugary drinks due to the WLs. CONCLUSION: Our findings suggest that implementation of WLs has reduced perceived purchases of unhealthy foods in Mexico. These results underscore the potential positive impact of the labeling policy particularly in subpopulations with lower levels of education and among indigenous adults.

Unsupervised domain adaptation with weak source domain labels via bidirectional subdomain alignment.

Unsupervised domain adaptation (UDA) enables knowledge transfer from a labeled source domain to an unlabeled target domain. However, UDA performance often relies heavily on the accuracy of source domain labels, which are frequently noisy or missing in real applications. To address unreliable source labels, we propose a novel framework for extracting robust, discriminative features via iterative pseudo-labeling, queue-based clustering, and bidirectional subdomain alignment (BSA). The proposed framework begins by generating pseudo-labels for unlabeled source data and constructing codebooks via iterative clustering to obtain label-independent class centroids. Then, the proposed framework performs two main tasks: rectifying features from both domains using BSA to match subdomain distributions and enhance features; and employing a two-stage adversarial process for global feature alignment. The feature rectification is done before feature enhancement, while the global alignment is done after feature enhancement. To optimize our framework, we formulate BSA and adversarial learning as maximizing a log-likelihood function, which is implemented via the Expectation-Maximization algorithm. The proposed framework shows significant improvements compared to state-of-the-art methods on Office-31, Office-Home, and VisDA-2017 datasets, achieving average accuracies of 91.5%, 76.6%, and 87.4%, respectively. Compared to existing methods, the proposed method shows consistent superiority in unsupervised domain adaptation tasks with both fully and weakly labeled source domains.

How Salty Is Too Salty? Designing Sodium Warning Label Policies to Identify High-Sodium Items on Restaurant Menus in the United States.

Two U.S. cities require chain restaurants to label menu items that exceed 100% of the Daily Value (DV) for sodium, informing consumers and potentially prompting restaurant reformulation. To inform policy design for other localities, this study determined the percentage of the top 91 U.S. chain restaurants' menu items that would be labeled if a warning policy were established for menu items exceeding the thresholds of 20%, 33%, 50%, 65%, and 100% of the sodium DV for adults. We obtained U.S. chain restaurants' nutrition information from the 2019 MenuStat database and calculated the percentage of items requiring sodium warning labels across the food and beverage categories at all the restaurants and at the full- and limited-service restaurants separately. In total, 19,038 items were included in the analyses. A warning label covering items with >20%, >33%, >50%, >65%, and >100% of the sodium DV resulted in expected coverage of 42%, 30%, 20%, 13%, and 5% of menu items at all the restaurants, respectively. At each threshold, the average percentage of items labeled per restaurant was higher among the full-service restaurants than the limited-service restaurants. These results suggest that restaurant warning policies with a threshold of 100% of the sodium DV per item would cover a minority of high-sodium menu items and that lower thresholds should be considered to help U.S. consumers reduce their sodium consumption.