Acoustic word embeddings (AWEs) are fixed-dimensional vector representations of speech segments that encode phonetic content so that different realisations of the same word have similar embeddings. In this paper we explore semantic AWE modelling. These AWEs should not only capture phonetics but also the meaning of a word (similar to textual word embeddings). We consider the scenario where we only have untranscribed speech in a target language. We introduce a number of strategies leveraging a pre-trained multilingual AWE model -- a phonetic AWE model trained on labelled data from multiple languages excluding the target. Our best semantic AWE approach involves clustering word segments using the multilingual AWE model, deriving soft pseudo-word labels from the cluster centroids, and then training a Skipgram-like model on the soft vectors. In an intrinsic word similarity task measuring semantics, this multilingual transfer approach outperforms all previous semantic AWE methods. We also show -- for the first time -- that AWEs can be used for downstream semantic query-by-example search.

Given a limited labeling budget, active learning (al) aims to sample the most informative instances from an unlabeled pool to acquire labels for subsequent model training. To achieve this, al typically measures the informativeness of unlabeled instances based on uncertainty and diversity. However, it does not consider erroneous instances with their neighborhood error density, which have great potential to improve the model performance. To address this limitation, we propose Real, a novel approach to select data instances with Representative Errors for Active Learning. It identifies minority predictions as pseudo errors within a cluster and allocates an adaptive sampling budget for the cluster based on estimated error density. Extensive experiments on five text classification datasets demonstrate that Real consistently outperforms all best-performing baselines regarding accuracy and F1-macro scores across a wide range of hyperparameter settings. Our analysis also shows that Real selects the most representative pseudo errors that match the distribution of ground-truth errors along the decision boundary.

In today's fast-paced world, the rates of stress and depression present a surge. Social media provide assistance for the early detection of mental health conditions. Existing methods mainly introduce feature extraction approaches and train shallow machine learning classifiers. Other researches use deep neural networks or transformers. Despite the fact that transformer-based models achieve noticeable improvements, they cannot often capture rich factual knowledge. Although there have been proposed a number of studies aiming to enhance the pretrained transformer-based models with extra information or additional modalities, no prior work has exploited these modifications for detecting stress and depression through social media. In addition, although the reliability of a machine learning model's confidence in its predictions is critical for high-risk applications, there is no prior work taken into consideration the model calibration. To resolve the above issues, we present the first study in the task of depression and stress detection in social media, which injects extra linguistic information in transformer-based models, namely BERT and MentalBERT. Specifically, the proposed approach employs a Multimodal Adaptation Gate for creating the combined embeddings, which are given as input to a BERT (or MentalBERT) model. For taking into account the model calibration, we apply label smoothing. We test our proposed approaches in three publicly available datasets and demonstrate that the integration of linguistic features into transformer-based models presents a surge in the performance. Also, the usage of label smoothing contributes to both the improvement of the model's performance and the calibration of the model. We finally perform a linguistic analysis of the posts and show differences in language between stressful and non-stressful texts, as well as depressive and non-depressive posts.

Language models (LMs) are pretrained on diverse data sources, including news, discussion forums, books, and online encyclopedias. A significant portion of this data includes opinions and perspectives which, on one hand, celebrate democracy and diversity of ideas, and on the other hand are inherently socially biased. Our work develops new methods to (1) measure political biases in LMs trained on such corpora, along social and economic axes, and (2) measure the fairness of downstream NLP models trained on top of politically biased LMs. We focus on hate speech and misinformation detection, aiming to empirically quantify the effects of political (social, economic) biases in pretraining data on the fairness of high-stakes social-oriented tasks. Our findings reveal that pretrained LMs do have political leanings that reinforce the polarization present in pretraining corpora, propagating social biases into hate speech predictions and misinformation detectors. We discuss the implications of our findings for NLP research and propose future directions to mitigate unfairness.

Given a huge, online stream of time-evolving events with multiple attributes, such as online shopping logs: (item, price, brand, time), and local mobility activities: (pick-up and drop-off locations, time), how can we summarize large, dynamic high-order tensor streams? How can we see any hidden patterns, rules, and anomalies? Our answer is to focus on two types of patterns, i.e., ''regimes'' and ''components'', for which we present CubeScope, an efficient and effective method over high-order tensor streams. Specifically, it identifies any sudden discontinuity and recognizes distinct dynamical patterns, ''regimes'' (e.g., weekday/weekend/holiday patterns). In each regime, it also performs multi-way summarization for all attributes (e.g., item, price, brand, and time) and discovers hidden ''components'' representing latent groups (e.g., item/brand groups) and their relationship. Thanks to its concise but effective summarization, CubeScope can also detect the sudden appearance of anomalies and identify the types of anomalies that occur in practice. Our proposed method has the following properties: (a) Effective: it captures dynamical multi-aspect patterns, i.e., regimes and components, and statistically summarizes all the events; (b) General: it is practical for successful application to data compression, pattern discovery, and anomaly detection on various types of tensor streams; (c) Scalable: our algorithm does not depend on the length of the data stream and its dimensionality. Extensive experiments on real datasets demonstrate that CubeScope finds meaningful patterns and anomalies correctly, and consistently outperforms the state-of-the-art methods as regards accuracy and execution speed.

In tasks like semantic parsing, instruction following, and question answering, standard deep networks fail to generalize compositionally from small datasets. Many existing approaches overcome this limitation with model architectures that enforce a compositional process of sentence interpretation. In this paper, we present a domain-general and model-agnostic formulation of compositionality as a constraint on symmetries of data distributions rather than models. Informally, we prove that whenever a task can be solved by a compositional model, there is a corresponding data augmentation scheme -- a procedure for transforming examples into other well formed examples -- that imparts compositional inductive bias on any model trained to solve the same task. We describe a procedure called LEXSYM that discovers these transformations automatically, then applies them to training data for ordinary neural sequence models. Unlike existing compositional data augmentation procedures, LEXSYM can be deployed agnostically across text, structured data, and even images. It matches or surpasses state-of-the-art, task-specific models on COGS semantic parsing, SCAN and ALCHEMY instruction following, and CLEVR-COGENT visual question answering datasets.

We present a deep learning method that propagates point-wise feature representations across shapes within a collection for the purpose of 3D shape segmentation. We propose a cross-shape attention mechanism to enable interactions between a shape's point-wise features and those of other shapes. The mechanism assesses both the degree of interaction between points and also mediates feature propagation across shapes, improving the accuracy and consistency of the resulting point-wise feature representations for shape segmentation. Our method also proposes a shape retrieval measure to select suitable shapes for cross-shape attention operations for each test shape. Our experiments demonstrate that our approach yields state-of-the-art results in the popular PartNet dataset.

Despite their remarkable success in approximating a wide range of operators defined by PDEs, existing neural operators (NOs) do not necessarily perform well for all physics problems. We focus here on high-frequency waves to highlight possible shortcomings. To resolve these, we propose a subfamily of NOs enabling an enhanced empirical approximation of the nonlinear operator mapping wave speed to solution, or boundary values for the Helmholtz equation on a bounded domain. The latter operator is commonly referred to as the ''forward'' operator in the study of inverse problems. Our methodology draws inspiration from transformers and techniques such as stochastic depth. Our experiments reveal certain surprises in the generalization and the relevance of introducing stochastic depth. Our NOs show superior performance as compared with standard NOs, not only for testing within the training distribution but also for out-of-distribution scenarios. To delve into this observation, we offer an in-depth analysis of the Rademacher complexity associated with our modified models and prove an upper bound tied to their stochastic depth that existing NOs do not satisfy. Furthermore, we obtain a novel out-of-distribution risk bound tailored to Gaussian measures on Banach spaces, again relating stochastic depth with the bound. We conclude by proposing a hypernetwork version of the subfamily of NOs as a surrogate model for the mentioned forward operator.

While they provide many applications, e.g., social robots, self-driving excellent prediction accuracy, their black-box vehicles, etc (Bennewitz, Burgard, & Thrun, 2002; nature makes it difficult for humans to interpret Thrun, Burgard, & Fox, 2005), and therefore the learned underlying function. Comparatively, has been studied in areas from computer science, model-based methods are based on explicit systems physics, and mathematics to robotics and transportation parameterized as ordinary/partial/stochastic (Bendali-Braham, Weber, Forestier, differentiable equations (O/P/SDEs) (Dietrich Idoumghar, & Muller, 2021). Existing research et al., 2021) or rule-based systems (Helbing & largely falls into model-free and model-based Molnár, 1995). These models are explainable but methods. Model-free methods enjoy the strong less accurate in prediction (Yue, Manocha, & data-fitting capacity of data-driven models such Wang, 2022), as they do not benefit from training as statistical machine learning models (Wang, on data (or only on small amounts of data) Ondřej, & O'Sullivan, 2016b; Wang & O'Sullivan, and therefore are better fit in small data regime.

The present study identifies and assesses the bibliographic trend in Artificial Intelligence (AI) research for the years 2015-2020 using the science mapping method of bibliometric study. The required data has been collected from the Scopus database. To make the collected data analysis-ready, essential data transformation was performed manually and with the help of a tool viz. OpenRefine. For determining the trend and performing the mapping techniques, top five open access and commercial journals of AI have been chosen based on their citescore driven ranking. The work includes 6880 articles published in the specified period for analysis. The trend is based on Country-wise publications, year-wise publications, topical terms in AI, top-cited articles, prominent authors, major institutions, involvement of industries in AI and Indian appearance. The results show that compared to open access journals; commercial journals have a higher citescore and number of articles published over the years. Additionally, IEEE is the prominent publisher which publishes 84% of the top-cited publications. Further, China and the United States are the major contributors to literature in the AI domain. The study reveals that neural networks and deep learning are the major topics included in top AI research publications. Recently, not only public institutions but also private bodies are investing their resources in AI research. The study also investigates the relative position of Indian researchers in terms of AI research. Present work helps in understanding the initial development, current stand and future direction of AI.