While the Large Language Models (LLMs) dominate a majority of language understanding tasks, previous work shows that some of these results are supported by modelling spurious correlations of training datasets.Authors commonly assess model robustness by evaluating their models on out-of-distribution (OOD) datasets of the same task, but these datasets might share the bias of the training dataset.We propose a simple method for measuring a scale of models' reliance on any identified spurious feature and assess the robustness towards a large set of known and newly found prediction biases for various pre-trained models and debiasing methods in Question Answering (QA).We find that the while existing debiasing methods can mitigate reliance on a chosen spurious feature, the OOD performance gains of these methods can not be explained by mitigated reliance on biased features, suggesting that biases are shared among different QA datasets.Finally, we evidence this to be the case by measuring that performance of models trained on different QA datasets rely on bias features comparably to the ID model.We hope these results will motivate future work to refine the reports of LMs' robustness to a level of adversarial samples addressing specific spurious features.

Domain adaptation allows generative language models to address specific flaws caused by the domain shift of their application.However, the traditional adaptation by further training on indomain data rapidly weakens the model's ability to generalize to other domains, making the openended deployments of the adapted models prone to errors.This work introduces novel training objectives built upon a semantic similarity of the predicted tokens to the reference.Our results show that (1) avoiding the common assumption of a single correct prediction by constructing the training target from tokens' semantic similarity can largely mitigate catastrophic forgetting of adaptation, while (2) preserving the adaptation in-domain quality, (3) with negligible additions to compute costs.In the broader context, the objectives grounded in a continuous token similarity pioneer the exploration of the middle ground between the efficient but naïve exact-match token-level objectives and expressive but computationally-and resourceintensive sequential objectives.

Which joint interactions in the human gait cycle can be used as biometric characteristics? Most current methods on gait recognition suffer from the lack of inter-pretability. We propose an interpretable feature representation of gait sequences by the graphical Granger causal inference. Gait sequence of a person in the standardized motion capture format, constituting a set of 3D joint spatial trajectories, is envisaged as a causal system of joints interacting in time. We apply the graphical Granger model (GGM) to obtain the so-called Granger causal graph among joints as a discriminative and visually interpretable representation of a person's gait. We evaluate eleven distance functions in the GGM feature space by established classification and class-separability evaluation metrics. Our experiments indicate that, depending on the metric, the most appropriate distance functions for the GGM are the total norm distance and the Ky-Fan 1-norm distance. Experiments also show that the GGM is able to detect the most discriminative joint interactions and that it outperforms five related interpretable models in correct classification rate and in Davies-Bouldin index. The proposed GGM model can serve as a complementary tool for gait analysis in kinesiology or for gait recognition in video surveillance.

Virtual reality has taken many great strides in the recent years. It is increasingly used and is an accepted means of delivering behavioral therapy for phobias and anxiety disorders. In this paper, we examine methods of virtual reality stress induction for use in treatment of somatoform and anxiety disorders, as well as the adequate measures of the evoked stress response. In total, 42 healthy subjects took part in testing as part of this study. The results show that electrodermal activity is more sensitive in capturing a subject reaction to non-interactive social environment simulation while cardiovascular parameters better reflect task-related stress. Furthermore, our results suggest a distinct relationship between electrodermal activity and anxiety and cardiovascular parameters and somatoform dissociation. These results can point to a possibility of virtual reality utilization in the research and treatment of disorders in which anxiety and somatization are important features.

Digital mathematical libraries (DMLs) such as arXiv, Numdam, and EuDML contain mainly documents from STEM fields, where mathematical formulae are often more important than text for understanding. Conventional information retrieval (IR) systems are unable to represent formulae and they are therefore ill-suited for math information retrieval (MIR). To fill the gap, we have developed, and open-sourced the MIaS MIR system. MIaS is based on the full-text search engine Apache Lucene. On top of text retrieval, MIaS also incorporates a set of tools for preprocessing mathematical formulae. We describe the design of the system and present speed, and quality evaluation results. We show that MIaS is both efficient, and effective, as evidenced by our victory in the NTCIR-11 Math-2 task.

Gait recognition from motion capture data, as a pattern classification discipline, can be improved by the use of machine learning. This article contributes to the state of the art with a statistical approach for extracting robust gait features directly from raw data by a modification of Linear Discriminant Analysis with Maximum Margin Criterion. Experiments on the CMU MoCap database show that the suggested method outperforms 13 relevant methods based on geometric features and a method to learn the features by a combination of Principal Component Analysis and Linear Discriminant Analysis. The methods are evaluated in terms of the distribution of biometric templates in respective feature spaces expressed in a number of class separability coefficients and classification metrics. Results also indicate a high portability of learned features, what means that we can learn what aspects of walk people generally differ in and extract those as general gait features. Recognizing people without needing group-specific features is convenient, as particular people might not always provide annotated learning data. As a contribution to reproducible research, our evaluation framework and database have been made publicly available. This research makes motion capture technology directly applicable for human recognition.

This work offers a design of a video surveillance system based on a soft biometric - gait identification from MoCap data. The main focus is on two substantial issues of the video surveillance scenario: (1) the walkers do not cooperate in providing learning data to establish their identities and (2) the data are often noisy or incomplete. We show that only a few examples of human gait cycles are required to learn a projection of raw MoCap data onto a low-dimensional subspace where the identities are well separable. Latent features learned by Maximum Margin Criterion (MMC) method discriminate better than any collection of geometric features. The MMC method is also highly robust to noisy data and works properly even with only a fraction of joints tracked. The overall workflow of the design is directly applicable for a day-to-day operation based on the available MoCap technology and algorithms for gait analysis. In the concept we introduce, a walker's identity is represented by a cluster of gait data collected at their incidents within the surveillance system: They are how they walk.

In the field of gait recognition from motion capture data, designing human-interpretable gait features is a common practice of many fellow researchers. To refrain from ad-hoc schemes and to find maximally discriminative features we may need to explore beyond the limits of human interpretability. This paper contributes to the state-of-the-art with a machine learning approach for extracting robust gait features directly from raw joint coordinates. The features are learned by a modification of Linear Discriminant Analysis with Maximum Margin Criterion so that the identities are maximally separated and, in combination with an appropriate classifier, used for gait recognition. Experiments on the CMU MoCap database show that this method outperforms eight other relevant methods in terms of the distribution of biometric templates in respective feature spaces expressed in four class separability coefficients. Additional experiments indicate that this method is a leading concept for rank-based classifier systems.

The design and architecture of MIaS (Math Indexer and Searcher), a system for mathematics retrieval is presented, and design decisions are discussed. We argue for an approach based on Presentation MathML using a similarity of math subformulae. The system was implemented as a math-aware search engine based on the state-of-the-art system Apache Lucene.

This paper describes several innovative document transformations and tools that have been developed in the process of building the Digital Mathematical Library DML-CZ http://dml.cz. The main result presented in this paper is our PDF re-compression tools developed using a jbig2enc library. Together with other programs, especially pdfsizeopt.py by Péter Szabó, we have managed to decrease PDF storage size and transmission needs be 62%: using both programs we reduced the size of the original PDFs to 38%.

This paper describes a technique to create interactive teaching materials as animations that are stored and distributed in PDF file format. PdfLATEX with small macropackage, Maple and Javascript are used and allow the development of interactive animations of high typographical quality that are fine-tuned for on-the-screen reading.

This paper describes a framework for effective design, typesetting, use and evaluation of students that uses multiple-choice tests. With this approach, based on a TEX engine, macros and a small program, several hundred customized tests can be typeset, printed and evaluated within several hours, allowing significant savings of educator's time and rapid electronic dissemination of tests results.

The use of JavaScript language for adding interaction to portable teaching materials of a high typographical quality in PDF file format is described. An extended version of the program TEX called pdfTEX is extremely useful for such purposes. It is shown that applications similar to those done by CGI script on the web can be done in PDF, exploiting the embedded JavaScript engine implementation in PDF viewers.