Search

The increasing consumption of news online in the 21st century coincided with increased publication of disinformation, biased reporting, hate speech and other unwanted Web content. We describe BiasScanner, an application that aims to strengthen democracy by supporting news consumers with scrutinizing news articles they are reading online. BiasScanner contains a server-side pre-trained large language model to identify biased sentences of news articles and a front-end Web browser plug-in. At the time of writing, BiasScanner can identify and classify more than two dozen types of media bias at the sentence level, making it the most fine-grained model and only deployed application (automatic system in use) of its kind. It was implemented in a light-weight and privacy-respecting manner, and in addition to highlighting likely biased sentence it also provides explanations for each classification decision as well as a summary analysis for each news article. While prior research has addressed news bias detection, we are not aware of any work that resulted in a deployed browser plug-in (c.f. also biasscanner.org for a Web demo)., Comment: 10 pages, 3 figures, 1 table

The World Wide Web provides unrivalled access to information globally, including factual news reporting and commentary. However, state actors and commercial players increasingly spread biased (distorted) or fake (non-factual) information to promote their agendas. We compare several large, pre-trained language models on the task of sentence-level news bias detection and sub-type classification, providing quantitative and qualitative results.

In this demonstration, we present Country Guesser, a live system that guesses the country that a photo is taken in. In particular, given a Google Street View image, our federated ranking model uses a combination of computer vision, machine learning and text retrieval methods to compute a ranking of likely countries of the location shown in a given image from Street View. Interestingly, using text-based features to probe large pre-trained language models can assist to provide cross-modal supervision. We are not aware of previous country guessing systems informed by visual and textual features.

Over the course of the recent decade, tremendous progress has been made in the areas of machine learning and natural language processing, which opened up vast areas of potential application use cases, including hiring and human resource management. We review the use cases for text analytics in the realm of human resources/personnel management, including actually realized as well as potential but not yet implemented ones, and we analyze the opportunities and risks of these., Comment: 10 pages, 2 figures, 1 table

Customer data typically is held in database systems, which can be seen as rule-based knowledge base, whereas businesses increasingly want to benefit from the capabilities of large, pre-trained language models. In this technical report, we describe a case study of how a commercial rule engine and an integrated neural chatbot may be integrated, and what level of control that particular integration mode leads to. We also discuss alternative ways (including past ways realized in other systems) how researchers strive to maintain control and avoid what has recently been called model "hallucination"., Comment: 12 pages, 3 figures

The rapid digitization of real-world data offers an unprecedented opportunity for optimizing healthcare delivery and accelerating biomedical discovery. In practice, however, such data is most abundantly available in unstructured forms, such as clinical notes in electronic medical records (EMRs), and it is generally plagued by confounders. In this paper, we present TRIALSCOPE, a unifying framework for distilling real-world evidence from population-level observational data. TRIALSCOPE leverages biomedical language models to structure clinical text at scale, employs advanced probabilistic modeling for denoising and imputation, and incorporates state-of-the-art causal inference techniques to combat common confounders. Using clinical trial specification as generic representation, TRIALSCOPE provides a turn-key solution to generate and reason with clinical hypotheses using observational data. In extensive experiments and analyses on a large-scale real-world dataset with over one million cancer patients from a large US healthcare network, we show that TRIALSCOPE can produce high-quality structuring of real-world data and generates comparable results to marquee cancer trials. In addition to facilitating in-silicon clinical trial design and optimization, TRIALSCOPE may be used to empower synthetic controls, pragmatic trials, post-market surveillance, as well as support fine-grained patient-like-me reasoning in precision diagnosis and treatment., Comment: 6 Figures, 22 Pages, 3 Tables

Humans have developed the capability to teach relevant aspects of new or adapted tasks to a social peer with very few task demonstrations by making use of scaffolding strategies that leverage prior knowledge and importantly prior joint experience to yield a joint understanding and a joint execution of the required steps to solve the task. This process has been discovered and analyzed in parent-infant interaction and constitutes a ``co-construction'' as it allows both, the teacher and the learner, to jointly contribute to the task. We propose to focus research in robot interactive learning on this co-construction process to enable robots to learn from non-expert users in everyday situations. In the following, we will review current proposals for interactive task learning and discuss their main contributions with respect to the entailing interaction. We then discuss our notion of co-construction and summarize research insights from adult-child and human-robot interactions to elucidate its nature in more detail. From this overview we finally derive research desiderata that entail the dimensions architecture, representation, interaction and explainability., Comment: 7 pages, 1 figure, accepted to ICRA 2023 Workshop 'Life-Long Learning with Human Help' (L3H2)

To meet the demands of jobs, now and in the future, jobseekers should possess a variety of skills frequently requested by employers. Skills can be both academic and technical in nature but also include personal qualities and advanced cognitive skills, such as decision-making and interpersonal skills--referred to in this report as success skills. Southern Regional Education Board (SREB) conducted a review of relevant research over the past 10 years (2013-2023) on non-technical skills to identify the success skills employers most often seek. This report summarizes relevant studies, skills gaps and credentialing to help policymakers, educators and businesses align efforts to prepare students for the workplace.

Big data, i.e. collecting, storing and processing of data at scale, has recently been possible due to the arrival of clusters of commodity computers powered by application-level distributed parallel operating systems like HDFS/Hadoop/Spark, and such infrastructures have revolutionized data mining at scale. For data mining project to succeed more consistently, some methodologies were developed (e.g. CRISP-DM, SEMMA, KDD), but these do not account for (1) very large scales of processing, (2) dealing with textual (unstructured) data (i.e. Natural Language Processing (NLP, "text analytics"), and (3) non-technical considerations (e.g. legal, ethical, project managerial aspects). To address these shortcomings, a new methodology, called "Data to Value" (D2V), is introduced, which is guided by a detailed catalog of questions in order to avoid a disconnect of big data text analytics project team with the topic when facing rather abstract box-and-arrow diagrams commonly associated with methodologies., Comment: 9 pages, 6 figures, 4 tables

In this paper we present irradiation measurements performed to select a transparent anode substrate that best meets the requirements of an optical readout for a novel detector, the LaGEMPix. The modification of the optical properties of the material due to proton irradiation were studied in soda-lime, fused quartz and fused silica glasses coated with an Indium Tin Oxide layer. The irradiations were performed using the research Beam Transfer Line (BTL) of the IBA Cyclone 18 MeV cyclotron of the Bern University Hospital (Inselspital). We recorded visible scintillation light generated by proton irradiation in the soda-lime and fused quartz samples. We also investigated the darkening of these three glasses and observed radiation-induced colour centres in the soda-lime glass sample. The optical transmission spectra of the samples were measured before and after irradiation. Reductions of 45%, 1% and 0.4% were observed for soda-lime glass, fused quartz and fused silica, respectively (with an associated error of 0.25%). We conclude that the best option for our specific application is the fused silica substrate, which will be the transparent anode for the next generation of the LaGEMPix detector., Comment: 13 pages, 10 Figures, Message from JINST: "We are pleased to inform you that your submission JINST_003T_0321 has been accepted for publication in JINST. [...]You are encouraged to post to arXiv"

Despite recent advances in deep learning-based language modelling, many natural language processing (NLP) tasks in the financial domain remain challenging due to the paucity of appropriately labelled data. Other issues that can limit task performance are differences in word distribution between the general corpora - typically used to pre-train language models - and financial corpora, which often exhibit specialized language and symbology. Here, we investigate two approaches that may help to mitigate these issues. Firstly, we experiment with further language model pre-training using large amounts of in-domain data from business and financial news. We then apply augmentation approaches to increase the size of our dataset for model fine-tuning. We report our findings on an Environmental, Social and Governance (ESG) controversies dataset and demonstrate that both approaches are beneficial to accuracy in classification tasks., Comment: 11 pages, 5 tables, 1 figure

We introduce Topic Grouper as a complementary approach in the field of probabilistic topic modeling. Topic Grouper creates a disjunctive partitioning of the training vocabulary in a stepwise manner such that resulting partitions represent topics. It is governed by a simple generative model, where the likelihood to generate the training documents via topics is optimized. The algorithm starts with one-word topics and joins two topics at every step. It therefore generates a solution for every desired number of topics ranging between the size of the training vocabulary and one. The process represents an agglomerative clustering that corresponds to a binary tree of topics. A resulting tree may act as a containment hierarchy, typically with more general topics towards the root of tree and more specific topics towards the leaves. Topic Grouper is not governed by a background distribution such as the Dirichlet and avoids hyper parameter optimizations. We show that Topic Grouper has reasonable predictive power and also a reasonable theoretical and practical complexity. Topic Grouper can deal well with stop words and function words and tends to push them into their own topics. Also, it can handle topic distributions, where some topics are more frequent than others. We present typical examples of computed topics from evaluation datasets, where topics appear conclusive and coherent. In this context, the fact that each word belongs to exactly one topic is not a major limitation; in some scenarios this can even be a genuine advantage, e.g.~a related shopping basket analysis may aid in optimizing groupings of articles in sales catalogs.

There has been a recent trend to migrate IT infrastructure into the cloud. In this paper, we discuss the impact of this trend on searching for textual and other data, i.e. the distributed indexing and retrieval of information, from an organizational context. Keywords: information retrieval (IR); federated search; cloud search., Comment: 6 pages, 1 figure, 1 table

Risk is part of the fabric of every business; surprisingly, there is little work on establishing best practices for systematic, repeatable risk identification, arguably the first step of any risk management process. In this paper, we present a proposal that constitutes a more holistic risk management approach, a methodology for computer-supported risk identification is proposed that may lead to more consistent (objective, repeatable) risk analysis., Comment: 19 pages, 5 figures

This tutorial presents a recipe for the construction of a compute cluster for processing large volumes of data, using cheap, easily available personal computer hardware (Intel/AMD based PCs) and freely available open source software (Ubuntu Linux, Apache Hadoop)., Comment: Technical Report; 15 pages, 1 figure

Biodiversity projections with uncertainty estimates under different climate, land-use, and policy scenarios are essential to setting and achieving international targets to mitigate biodiversity loss. Evaluating and improving biodiversity predictions to better inform policy decisions remains a central conservation goal and challenge. A comprehensive strategy to evaluate and reduce uncertainty of model outputs against observed measurements and multiple models would help to produce more robust biodiversity predictions. We propose an approach that integrates biodiversity models and emerging remote sensing and in-situ data streams to evaluate and reduce uncertainty with the goal of improving policy-relevant biodiversity predictions. In this article, we describe a multivariate approach to directly and indirectly evaluate and constrain model uncertainty, demonstrate a proof of concept of this approach, embed the concept within the broader context of model evaluation and scenario analysis for conservation policy, and highlight lessons from other modeling communities.

Information needs are naturally represented as questions. Automatic Natural-Language Question Answering (NLQA) has only recently become a practical task on a larger scale and without domain constraints. This paper gives a brief introduction to the field, its history and the impact of systematic evaluation competitions. It is then demonstrated that an NLQA system for English can be built and evaluated in a very short time using off-the-shelf parsers and thesauri. The system is based on Robust Minimal Recursion Semantics (RMRS) and is portable with respect to the parser used as a frontend. It applies atomic term unification supported by question classification and WordNet lookup for semantic similarity matching of parsed question representation and free text., Comment: 8 pages, 6 figures, 5 tables. To appear in Proc. TaCoS'02, Potsdam, Germany

The U.S. Global Change Research Program Sustained Assessment process benefits from long-term investments in Earth science research that enable the scientific community to conduct assessment-relevant science. To this end, NASA initiated several research programs over the past five years to support the Earth observation community in developing indicators, datasets, research products, and tools to support ongoing and future National Climate Assessments. These activities complement NASA's ongoing Earth science research programs. One aspect of the assessment portfolio funds four "enabling tools" projects at NASA research centers. Each tool leverages existing capacity within the center, but has developed tailored applications and products for National Climate Assessments. The four projects build on the capabilities of a global atmospheric reanalysis (MERRA-2), a continental U.S. land surface reanalysis (NCA-LDAS), the NASA Earth Exchange (NEX), and a Regional Climate Model Evaluation System (RCMES). Here, we provide a brief overview of each enabling tool, highlighting the ways in which it has advanced assessment science to date. We also discuss how the assessment community can access and utilize these tools for National Climate Assessments and other sustained assessment activities.

In this study, we apply a two-dimensional variational analysis method (2d-VAR) to select a wind solution from NASA Scatterometer (NSCAT) ambiguous winds. 2d-VAR determines a "best" gridded surface wind analysis by minimizing a cost function. The cost function measures the misfit to the observations, the background, and the filtering and dynamical constraints. The ambiguity closest in direction to the minimizing analysis is selected. 2d-VAR method, sensitivity and numerical behavior are described. 2d-VAR is compared to statistical interpolation (OI) by examining the response of both systems to a single ship observation and to a swath of unique scatterometer winds. 2d-VAR is used with both NSCAT ambiguities and NSCAT backscatter values. Results are roughly comparable. When the background field is poor, 2d-VAR ambiguity removal often selects low probability ambiguities. To avoid this behavior, an initial 2d-VAR analysis, using only the two most likely ambiguities, provides the first guess for an analysis using all the ambiguities or the backscatter data. 2d-VAR and median filter selected ambiguities usually agree. Both methods require horizontal consistency, so disagreements occur in clumps, or as linear features. In these cases, 2d-VAR ambiguities are often more meteorologically reasonable and more consistent with satellite imagery.