Search

Laboratory accidents pose significant risks to human life and property, underscoring the importance of robust safety protocols. Despite advancements in safety training, laboratory personnel may still unknowingly engage in unsafe practices. With the increasing reliance on large language models (LLMs) for guidance in various fields, including laboratory settings, there is a growing concern about their reliability in critical safety-related decision-making. Unlike trained human researchers, LLMs lack formal lab safety education, raising questions about their ability to provide safe and accurate guidance. Existing research on LLM trustworthiness primarily focuses on issues such as ethical compliance, truthfulness, and fairness but fails to fully cover safety-critical real-world applications, like lab safety. To address this gap, we propose the Laboratory Safety Benchmark (LabSafety Bench), a comprehensive evaluation framework based on a new taxonomy aligned with Occupational Safety and Health Administration (OSHA) protocols. This benchmark includes 765 multiple-choice questions verified by human experts, assessing LLMs and vision language models (VLMs) performance in lab safety contexts. Our evaluations demonstrate that while GPT-4o outperforms human participants, it is still prone to critical errors, highlighting the risks of relying on LLMs in safety-critical environments. Our findings emphasize the need for specialized benchmarks to accurately assess the trustworthiness of LLMs in real-world safety applications., Comment: 50 pages, 19 figures

Large language models (LLMs) offer powerful capabilities but also introduce significant risks. One way to mitigate these risks is through comprehensive pre-deployment evaluations using benchmarks designed to test for specific vulnerabilities. However, the rapidly expanding body of LLM benchmark literature lacks a standardized method for documenting crucial benchmark details, hindering consistent use and informed selection. BenchmarkCards addresses this gap by providing a structured framework specifically for documenting LLM benchmark properties rather than defining the entire evaluation process itself. BenchmarkCards do not prescribe how to measure or interpret benchmark results (e.g., defining ``correctness'') but instead offer a standardized way to capture and report critical characteristics like targeted risks and evaluation methodologies, including properties such as bias and fairness. This structured metadata facilitates informed benchmark selection, enabling researchers to choose appropriate benchmarks and promoting transparency and reproducibility in LLM evaluation.

LLM-as-a-Judge has been widely utilized as an evaluation method in various benchmarks and served as supervised rewards in model training. However, despite their excellence in many domains, potential issues are under-explored, undermining their reliability and the scope of their utility. Therefore, we identify 12 key potential biases and propose a new automated bias quantification framework-CALM-which systematically quantifies and analyzes each type of bias in LLM-as-a-Judge by using automated and principle-guided modification. Our experiments cover multiple popular language models, and the results indicate that while advanced models have achieved commendable overall performance, significant biases persist in certain specific tasks. Empirical results suggest that there remains room for improvement in the reliability of LLM-as-a-Judge. Moreover, we also discuss the explicit and implicit influence of these biases and give some suggestions for the reliable application of LLM-as-a-Judge. Our work highlights the need for stakeholders to address these issues and remind users to exercise caution in LLM-as-a-Judge applications.

The Covid-19 pandemic has affected the world at multiple levels. Data sharing was pivotal for advancing research to understand the underlying causes and implement effective containment strategies. In response, many countries have promoted the availability of daily cases to support research initiatives, fostering collaboration between organisations and making such data available to the public through open data platforms. Despite the several advantages of data sharing, one of the major concerns before releasing health data is its impact on individuals' privacy. Such a sharing process should be based on state-of-the-art methods in Data Protection by Design and by Default. In this paper, we use a data set related to Covid-19 cases in the second largest hospital in Portugal to show how it is feasible to ensure data privacy while improving the quality and maintaining the utility of the data. Our goal is to demonstrate how knowledge exchange in multidisciplinary teams of healthcare practitioners, data privacy, and data science experts is crucial to co-developing strategies that ensure high utility of de-identified data., Comment: 7 figures and 8 tables

Sharing private data for learning tasks is pivotal for transparent and secure machine learning applications. Many privacy-preserving techniques have been proposed for this task aiming to transform the data while ensuring the privacy of individuals. Some of these techniques have been incorporated into tools, whereas others are accessed through various online platforms. However, such tools require manual configuration, which can be complex and time-consuming. Moreover, they require substantial expertise, potentially restricting their use to those with advanced technical knowledge. In this paper, we propose AUTOPRIV, the first automated privacy-preservation method, that eliminates the need for any manual configuration. AUTOPRIV employs meta-learning to automate the de-identification process, facilitating the secure release of data for machine learning tasks. The main goal is to anticipate the predictive performance and privacy risk of a large set of privacy configurations. We provide a ranked list of the most promising solutions, which are likely to achieve an optimal approximation within a new domain. AUTOPRIV is highly effective as it reduces computational complexity and energy consumption considerably., Comment: 12 pages, 6 figures, 3 tables

Multiple synthetic data generation models have emerged, among which deep learning models have become the vanguard due to their ability to capture the underlying characteristics of the original data. However, the resemblance of the synthetic to the original data raises important questions on the protection of individuals' privacy. As synthetic data is perceived as a means to fully protect personal information, most current related work disregards the impact of re-identification risk. In particular, limited attention has been given to exploring outliers, despite their privacy relevance. In this work, we analyze the privacy of synthetic data w.r.t the outliers. Our main findings suggest that outliers re-identification via linkage attack is feasible and easily achieved. Furthermore, additional safeguards such as differential privacy can prevent re-identification, albeit at the expense of the data utility.

As machine learning continues to gain prominence, transparency and explainability are increasingly critical. Without an understanding of these models, they can replicate and worsen human bias, adversely affecting marginalized communities. Algorithmic recourse emerges as a tool for clarifying decisions made by predictive models, providing actionable insights to alter outcomes. They answer, 'What do I have to change?' to achieve the desired result. Despite their importance, current algorithmic recourse methods treat all domain values equally, which is unrealistic in real-world settings. In this paper, we propose a novel framework, Relevance-Aware Algorithmic Recourse (RAAR), that leverages the concept of relevance in applying algorithmic recourse to regression tasks. We conducted multiple experiments on 15 datasets to outline how relevance influences recourses. Results show that relevance contributes algorithmic recourses comparable to well-known baselines, with greater efficiency and lower relative costs., Comment: 5 pages (4 content, 1 references)

A long-standing dilemma prevents the broader application of explanation methods: general applicability and inference speed. On the one hand, existing model-agnostic explanation methods usually make minimal pre-assumptions about the prediction models to be explained. Still, they require additional queries to the model through propagation or back-propagation to approximate the models' behaviors, resulting in slow inference and hindering their use in time-sensitive tasks. On the other hand, various model-dependent explanations have been proposed that achieve low-cost, fast inference but at the expense of limiting their applicability to specific model structures. In this study, we bridge the gap between the universality of model-agnostic approaches and the efficiency of model-specific approaches by proposing a novel framework without assumptions on the prediction model's structures, achieving high efficiency during inference and allowing for real-time explanations. To achieve this, we first define explanations through a set of human-comprehensible concepts and propose a framework to elucidate model predictions via minimal feasible concept sets. Second, we show that a minimal feasible set generator can be learned as a companion explainer to the prediction model, generating explanations for predictions. Finally, we validate this framework by implementing a novel model-agnostic method that provides robust explanations while facilitating real-time inference. Our claims are substantiated by comprehensive experiments, highlighting the effectiveness and efficiency of our approach.

Many evaluation metrics can be used to assess the performance of models in binary classification tasks. However, most of them are derived from a confusion matrix in a non-differentiable form, making it very difficult to generate a differentiable loss function that could directly optimize them. The lack of solutions to bridge this challenge not only hinders our ability to solve difficult tasks, such as imbalanced learning, but also requires the deployment of computationally expensive hyperparameter search processes in model selection. In this paper, we propose a general-purpose approach that transforms any confusion matrix-based metric into a loss function, \textit{AnyLoss}, that is available in optimization processes. To this end, we use an approximation function to make a confusion matrix represented in a differentiable form, and this approach enables any confusion matrix-based metric to be directly used as a loss function. The mechanism of the approximation function is provided to ensure its operability and the differentiability of our loss functions is proved by suggesting their derivatives. We conduct extensive experiments under diverse neural networks with many datasets, and we demonstrate their general availability to target any confusion matrix-based metrics. Our method, especially, shows outstanding achievements in dealing with imbalanced datasets, and its competitive learning speed, compared to multiple baseline models, underscores its efficiency.

Recent state-of-the-art forecasting methods are trained on collections of time series. These methods, often referred to as global models, can capture common patterns in different time series to improve their generalization performance. However, they require large amounts of data that might not be readily available. Besides this, global models sometimes fail to capture relevant patterns unique to a particular time series. In these cases, data augmentation can be useful to increase the sample size of time series datasets. The main contribution of this work is a novel method for generating univariate time series synthetic samples. Our approach stems from the insight that the observations concerning a particular time series of interest represent only a small fraction of all observations. In this context, we frame the problem of training a forecasting model as an imbalanced learning task. Oversampling strategies are popular approaches used to deal with the imbalance problem in machine learning. We use these techniques to create synthetic time series observations and improve the accuracy of forecasting models. We carried out experiments using 7 different databases that contain a total of 5502 univariate time series. We found that the proposed solution outperforms both a global and a local model, thus providing a better trade-off between these two approaches.

Should prediction models always deliver a prediction? In the pursuit of maximum predictive performance, critical considerations of reliability and fairness are often overshadowed, particularly when it comes to the role of uncertainty. Selective regression, also known as the "reject option," allows models to abstain from predictions in cases of considerable uncertainty. Initially proposed seven decades ago, approaches to selective regression have mostly focused on distribution-based proxies for measuring uncertainty, particularly conditional variance. However, this focus neglects the significant influence of model-specific biases on a model's performance. In this paper, we propose a novel approach to selective regression by leveraging conformal prediction, which provides grounded confidence measures for individual predictions based on model-specific biases. In addition, we propose a standardized evaluation framework to allow proper comparison of selective regression approaches. Via an extensive experimental approach, we demonstrate how our proposed approach, conformalized selective regression, demonstrates an advantage over multiple state-of-the-art baselines.

The yield of a chemical reaction quantifies the percentage of the target product formed in relation to the reactants consumed during the chemical reaction. Accurate yield prediction can guide chemists toward selecting high-yield reactions during synthesis planning, offering valuable insights before dedicating time and resources to wet lab experiments. While recent advancements in yield prediction have led to overall performance improvement across the entire yield range, an open challenge remains in enhancing predictions for high-yield reactions, which are of greater concern to chemists. In this paper, we argue that the performance gap in high-yield predictions results from the imbalanced distribution of real-world data skewed towards low-yield reactions, often due to unreacted starting materials and inherent ambiguities in the reaction processes. Despite this data imbalance, existing yield prediction methods continue to treat different yield ranges equally, assuming a balanced training distribution. Through extensive experiments on three real-world yield prediction datasets, we emphasize the urgent need to reframe reaction yield prediction as an imbalanced regression problem. Finally, we demonstrate that incorporating simple cost-sensitive re-weighting methods can significantly enhance the performance of yield prediction models on underrepresented high-yield regions.

In the context of the global seafood industry, the Azores archipelago (Portugal) plays a pivotal role due to its vast maritime domain. This study employs complex network analysis techniques to investigate the dynamics of Azores fisheries, using time series data converted into networks. We uncover associations between Tunas and specific islands, consistent links among fish classifications, and identify other pivotal nodes within the fishing network. Remarkably, nodes with high degrees and a local clustering coefficient of one provide crucial insights into the fishing ecosystem. This study highlights the value of network analysis for understanding fisheries complexities and offers insights into sustainable management and the preservation of marine ecosystems. It also emphasizes the urgency for ongoing research and data collection to enrich our understanding of this multifaceted domain.

Fishery analysis is critical in maintaining the long-term sustainability of species and the livelihoods of millions of people who depend on fishing for food and income. The fishing gear, or metier, is a key factor significantly impacting marine habitats, selectively targeting species and fish sizes. Analysis of commercial catches or landings by metier in fishery stock assessment and management is crucial, providing robust estimates of fishing efforts and their impact on marine ecosystems. In this paper, we focus on a unique data set from the Azores' fishing data collection programs between 2010 and 2017, where little information on metiers is available and sparse throughout our timeline. Our main objective is to tackle the task of data set reconstruction, leveraging domain knowledge and machine learning methods to retrieve or associate metier-related information to each fish landing. We empirically validate the feasibility of this task using a diverse set of modeling approaches and demonstrate how it provides new insights into different fisheries' behavior and the impact of metiers over time, which are essential for future fish population assessments, management, and conservation efforts.

As the frontier of machine learning applications moves further into human interaction, multiple concerns arise regarding automated decision-making. Two of the most critical issues are fairness and data privacy. On the one hand, one must guarantee that automated decisions are not biased against certain groups, especially those unprotected or marginalized. On the other hand, one must ensure that the use of personal information fully abides by privacy regulations and that user identities are kept safe. The balance between privacy, fairness, and predictive performance is complex. However, despite their potential societal impact, we still demonstrate a poor understanding of the dynamics between these optimization vectors. In this paper, we study this three-way tension and how the optimization of each vector impacts others, aiming to inform the future development of safe applications. In light of claims that predictive performance and fairness can be jointly optimized, we find this is only possible at the expense of data privacy. Overall, experimental results show that one of the vectors will be penalized regardless of which of the three we optimize. Nonetheless, we find promising avenues for future work in joint optimization solutions, where smaller trade-offs are observed between the three vectors., Comment: 12, 6 figures and 2 tables

The rapid advancement in data-driven research has increased the demand for effective graph data analysis. However, real-world data often exhibits class imbalance, leading to poor performance of machine learning models. To overcome this challenge, class-imbalanced learning on graphs (CILG) has emerged as a promising solution that combines the strengths of graph representation learning and class-imbalanced learning. In recent years, significant progress has been made in CILG. Anticipating that such a trend will continue, this survey aims to offer a comprehensive understanding of the current state-of-the-art in CILG and provide insights for future research directions. Concerning the former, we introduce the first taxonomy of existing work and its connection to existing imbalanced learning literature. Concerning the latter, we critically analyze recent work in CILG and discuss urgent lines of inquiry within the topic. Moreover, we provide a continuously maintained reading list of papers and code at https://github.com/yihongma/CILG-Papers., Comment: submitted to ACM Computing Survey (CSUR)

In natural phenomena, data distributions often deviate from normality. One can think of cataclysms as a self-explanatory example: events that occur almost never, and at the same time are many standard deviations away from the common outcome. In many scientific contexts it is exactly these tail events that researchers are most interested in anticipating, so that adequate measures can be taken to prevent or attenuate a major impact on society. Despite such efforts, we have yet to provide definite answers to crucial issues in evaluating predictive solutions in domains such as weather, pollution, health. In this paper, we deal with two encapsulated problems simultaneously. First, assessing the performance of regression models when non-uniform preferences apply - not all values are equally relevant concerning the accuracy of their prediction, and there's a particular interest in the most extreme values. Second, assessing the robustness of models when dealing with uncertainty regarding the actual underlying distribution of values relevant for such problems. We show how different levels of relevance associated with target values may impact experimental conclusions, and demonstrate the practical utility of the proposed methods.

Protecting user data privacy can be achieved via many methods, from statistical transformations to generative models. However, all of them have critical drawbacks. For example, creating a transformed data set using traditional techniques is highly time-consuming. Also, recent deep learning-based solutions require significant computational resources in addition to long training phases, and differentially private-based solutions may undermine data utility. In this paper, we propose $\epsilon$-PrivateSMOTE, a technique designed for safeguarding against re-identification and linkage attacks, particularly addressing cases with a high \sloppy re-identification risk. Our proposal combines synthetic data generation via noise-induced interpolation with differential privacy principles to obfuscate high-risk cases. We demonstrate how $\epsilon$-PrivateSMOTE is capable of achieving competitive results in privacy risk and better predictive performance when compared to multiple traditional and state-of-the-art privacy-preservation methods, including generative adversarial networks, variational autoencoders, and differential privacy baselines. We also show how our method improves time requirements by at least a factor of 9 and is a resource-efficient solution that ensures high performance without specialised hardware., Comment: 21 pages, 6 figures and 2 tables

Imbalanced domain learning aims to produce accurate models in predicting instances that, though underrepresented, are of utmost importance for the domain. Research in this field has been mainly focused on classification tasks. Comparatively, the number of studies carried out in the context of regression tasks is negligible. One of the main reasons for this is the lack of loss functions capable of focusing on minimizing the errors of extreme (rare) values. Recently, an evaluation metric was introduced: Squared Error Relevance Area (SERA). This metric posits a bigger emphasis on the errors committed at extreme values while also accounting for the performance in the overall target variable domain, thus preventing severe bias. However, its effectiveness as an optimization metric is unknown. In this paper, our goal is to study the impacts of using SERA as an optimization criterion in imbalanced regression tasks. Using gradient boosting algorithms as proof of concept, we perform an experimental study with 36 data sets of different domains and sizes. Results show that models that used SERA as an objective function are practically better than the models produced by their respective standard boosting algorithms at the prediction of extreme values. This confirms that SERA can be embedded as a loss function into optimization-based learning algorithms for imbalanced regression scenarios.

The exponential growth of collected, processed, and shared microdata has given rise to concerns about individuals' privacy. As a result, laws and regulations have emerged to control what organisations do with microdata and how they protect it. Statistical Disclosure Control seeks to reduce the risk of confidential information disclosure by de-identifying them. Such de-identification is guaranteed through privacy-preserving techniques. However, de-identified data usually results in loss of information, with a possible impact on data analysis precision and model predictive performance. The main goal is to protect the individuals' privacy while maintaining the interpretability of the data, i.e. its usefulness. Statistical Disclosure Control is an area that is expanding and needs to be explored since there is still no solution that guarantees optimal privacy and utility. This survey focuses on all steps of the de-identification process. We present existing privacy-preserving techniques used in microdata de-identification, privacy measures suitable for several disclosure types and, information loss and predictive performance measures. In this survey, we discuss the main challenges raised by privacy constraints, describe the main approaches to handle these obstacles, review taxonomies of privacy-preserving techniques, provide a theoretical analysis of existing comparative studies, and raise multiple open issues., Comment: 35 pages, 3 figures, 5 tables

Machine learning is increasingly used in the most diverse applications and domains, whether in healthcare, to predict pathologies, or in the financial sector to detect fraud. One of the linchpins for efficiency and accuracy in machine learning is data utility. However, when it contains personal information, full access may be restricted due to laws and regulations aiming to protect individuals' privacy. Therefore, data owners must ensure that any data shared guarantees such privacy. Removal or transformation of private information (de-identification) are among the most common techniques. Intuitively, one can anticipate that reducing detail or distorting information would result in losses for model predictive performance. However, previous work concerning classification tasks using de-identified data generally demonstrates that predictive performance can be preserved in specific applications. In this paper, we aim to evaluate the existence of a trade-off between data privacy and predictive performance in classification tasks. We leverage a large set of privacy-preserving techniques and learning algorithms to provide an assessment of re-identification ability and the impact of transformed variants on predictive performance. Unlike previous literature, we confirm that the higher the level of privacy (lower re-identification risk), the higher the impact on predictive performance, pointing towards clear evidence of a trade-off., Comment: 29 pages, 9 figures, 3 tables

Time series forecasting is a challenging task with applications in a wide range of domains. Auto-regression is one of the most common approaches to address these problems. Accordingly, observations are modelled by multiple regression using their past lags as predictor variables. We investigate the extension of auto-regressive processes using statistics which summarise the recent past dynamics of time series. The result of our research is a novel framework called VEST, designed to perform feature engineering using univariate and numeric time series automatically. The proposed approach works in three main steps. First, recent observations are mapped onto different representations. Second, each representation is summarised by statistical functions. Finally, a filter is applied for feature selection. We discovered that combining the features generated by VEST with auto-regression significantly improves forecasting performance. We provide evidence using 90 time series with high sampling frequency. VEST is publicly available online., Comment: 25 pages, R code available

This paper presents a data set describing the evolution of results in the Portuguese Parliamentary Elections of October 6$^{th}$ 2019. The data spans a time interval of 4 hours and 25 minutes, in intervals of 5 minutes, concerning the results of the 27 parties involved in the electoral event. The data set is tailored for predictive modelling tasks, mostly focused on numerical forecasting tasks. Regardless, it allows for other tasks such as ordinal regression or learn-to-rank., Comment: Dataset Descriptor

The profusion of user generated content caused by the rise of social media platforms has enabled a surge in research relating to fields such as information retrieval, recommender systems, data mining and machine learning. However, the lack of comprehensive baseline data sets to allow a thorough evaluative comparison has become an important issue. In this paper we present a large data set of news items from well-known aggregators such as Google News and Yahoo! News, and their respective social feedback on multiple platforms: Facebook, Google+ and LinkedIn. The data collected relates to a period of 8 months, between November 2015 and July 2016, accounting for about 100,000 news items on four different topics: economy, microsoft, obama and palestine. This data set is tailored for evaluative comparisons in predictive analytics tasks, although allowing for tasks in other research areas such as topic detection and tracking, sentiment analysis in short text, first story detection or news recommendation., Comment: Dataset available in http://www.dcc.fc.up.pt/~nmoniz/MultiSourceNews

Ranking evaluation metrics are a fundamental element of design and improvement efforts in information retrieval. We observe that most popular metrics disregard information portrayed in the scores used to derive rankings, when available. This may pose a numerical scaling problem, causing an under- or over-estimation of the evaluation depending on the degree of divergence between the scores of ranked items. The purpose of this work is to propose a principled way of quantifying multi-graded relevance judgments of items and enable a more accurate penalization of ordering errors in rankings. We propose a data-driven generation of relevance functions based on the degree of the divergence amongst a set of items' scores and its application in the evaluation metric Normalized Discounted Cumulative Gain (nDCG). We use synthetic data to demonstrate the interest of our proposal and a combination of data on news items from Google News and their respective popularity in Twitter to show its performance in comparison to the standard nDCG. Results show that our proposal is capable of providing a more fine-grained evaluation of rankings when compared to the standard nDCG, and that the latter frequently under- or over-estimates its evaluation scores in light of the divergence of items' scores., Comment: 12 pages, 4 figures

The participatory Web has enabled the ubiquitous and pervasive access of information, accompanied by an increase of speed and reach in information sharing. Data dissemination services such as news aggregators are expected to provide up-to-date, real-time information to the end users. News aggregators are in essence recommendation systems that filter and rank news stories in order to select the few that will appear on the users front screen at any time. One of the main challenges in such systems is to address the recency and latency problems, that is, to identify as soon as possible how important a news story is. In this work we propose an integrated framework that aims at predicting the importance of news items upon their publication with a focus on recent and highly popular news, employing resampling strategies, and at translating the result into concrete news rankings. We perform an extensive experimental evaluation using real-life datasets of the proposed framework as both a stand-alone system and when applied to news recommendations from Google News. Additionally, we propose and evaluate a combinatorial solution to the augmentation of official media recommendations with social information. Results show that the proposed approach complements and enhances the news rankings generated by state-of-the-art systems., Comment: 17 pages, 2 figures, submitted to the ACM Transactions on Intelligent Systems and Technology (ACM TIST), Special Issue on Social Media Processing