Search

In spite of the recent progress in speech processing, the majority of world languages and dialects remain uncovered. This situation only furthers an already wide technological divide, thereby hindering technological and socioeconomic inclusion. This challenge is largely due to the absence of datasets that can empower diverse speech systems. In this paper, we seek to mitigate this obstacle for a number of Arabic dialects by presenting Casablanca, a large-scale community-driven effort to collect and transcribe a multi-dialectal Arabic dataset. The dataset covers eight dialects: Algerian, Egyptian, Emirati, Jordanian, Mauritanian, Moroccan, Palestinian, and Yemeni, and includes annotations for transcription, gender, dialect, and code-switching. We also develop a number of strong baselines exploiting Casablanca. The project page for Casablanca is accessible at: www.dlnlp.ai/speech/casablanca.

An optimal microswimmer with a given geometry has a surface velocity profile that minimizes energy dissipation for a given swimming speed. An axisymmetric swimmer can be puller-, pusher-, or neutral-type depending on the sign of the stresslet strength. We numerically investigate the type of optimal surface-driven active microswimmers using a minimum dissipation theorem for optimum microswimmers. We examine the hydrodynamic resistance and stresslet strength with nonlinear dependence on various deformation modes. The results show that microswimmers with fore-and-aft symmetry exhibit neutral-type behavior, while asymmetrical geometries exhibit pusher-, puller-type or neutral-type behavior, depending on the dominant deformation mode and the nonlinear dependence of the stresslet for an optimum microswimmer on deformation mode and amplitude., Comment: 10 pages, 3 figures

Determining the crossing numbers of Cartesian products of small graphs with arbitrarily large paths has been an ongoing topic of research since the 1970s. Doing so requires the establishment of coincident upper and lower bounds; the former is usually demonstrated by providing a suitable drawing procedure, while the latter often requires substantial theoretical arguments. Many such papers have been published, which typically focus on just one or two small graphs at a time, and use ad hoc arguments specific to those graphs. We propose a general approach which, when successful, establishes the required lower bound. This approach can be applied to the Cartesian product of any graph with arbitrarily large paths, and in each case involves solving a modified version of the crossing number problem on a finite number (typically only two or three) of small graphs. We demonstrate the potency of this approach by applying it to Cartesian products involving all 133 graphs $G$ of orders five or six, and show that it is successful in 128 cases. This includes 60 cases which a recent survey listed as either undetermined, or determined only in journals without adequate peer review.

Quantum magnetic materials can provide explicit realizations of paradigm models in quantum many-body physics. In this context, SrCu$_2$(BO$_3$)$_2$ is a faithful realization of the Shastry-Sutherland model (SSM) for ideally frustrated spin dimers, even displaying several of its quantum magnetic phases as a function of pressure. We perform inelastic neutron scattering (INS) measurements on SrCu$_2$(BO$_3$)$_2$ at 5.5 GPa and 4.5 K, observing spin waves that characterize the high-pressure antiferromagnetic phase. The experimental spectra are well described by linear spin-wave calculations on a SSM with an inter-layer interaction, which is determined accurately as $J_c = 0.053(3)$ meV. The presence of $J_c$ indicates the need to account for the three-dimensional nature of SrCu$_2$(BO$_3$)$_2$ in theoretical models, also at lower pressures. We find that the ratio between in-plane interactions, $J'/J = 1.8(2)$, undergoes a dramatic change compared to lower pressures that we deduce is driven by a sharp drop in the dimer coupling, $J$. Our results underline the wide horizons opened by high-pressure INS experiments on quantum magnetic materials., Comment: Main text + supplemental material

The widespread use of large language models has brought up essential questions about the potential biases these models might learn. This led to the development of several metrics aimed at evaluating and mitigating these biases. In this paper, we first demonstrate that prompt-based fairness metrics exhibit poor agreement, as measured by correlation, raising important questions about the reliability of fairness assessment using prompts. Then, we outline six relevant reasons why such a low correlation is observed across existing metrics. Based on these insights, we propose a method called Correlated Fairness Output (CAIRO) to enhance the correlation between fairness metrics. CAIRO augments the original prompts of a given fairness metric by using several pre-trained language models and then selects the combination of the augmented prompts that achieves the highest correlation across metrics. We show a significant improvement in Pearson correlation from 0.3 and 0.18 to 0.90 and 0.98 across metrics for gender and religion biases, respectively. Our code is available at https://github.com/chandar-lab/CAIRO., Comment: In Proceedings of ACL main 2024

Aspect-Based Sentiment Analysis (ABSA) aims to identify terms or multiword expressions (MWEs) on which sentiments are expressed and the sentiment polarities associated with them. The development of supervised models has been at the forefront of research in this area. However, training these models requires the availability of manually annotated datasets which is both expensive and time-consuming. Furthermore, the available annotated datasets are tailored to a specific domain, language, and text type. In this work, we address this notable challenge in current state-of-the-art ABSA research. We propose a hybrid approach for Aspect Based Sentiment Analysis using transfer learning. The approach focuses on generating weakly-supervised annotations by exploiting the strengths of both large language models (LLM) and traditional syntactic dependencies. We utilise syntactic dependency structures of sentences to complement the annotations generated by LLMs, as they may overlook domain-specific aspect terms. Extensive experimentation on multiple datasets is performed to demonstrate the efficacy of our hybrid method for the tasks of aspect term extraction and aspect sentiment classification. Keywords: Aspect Based Sentiment Analysis, Syntactic Parsing, large language model (LLM)

The construction industry is a vital sector of the global economy, but it faces many productivity challenges in various processes, such as design, planning, procurement, inspection, and maintenance. Generative artificial intelligence (AI), which can create novel and realistic data or content, such as text, image, video, or code, based on some input or prior knowledge, offers innovative and disruptive solutions to address these challenges. However, there is a gap in the literature on the current state, opportunities, and challenges of generative AI in the construction industry. This study aims to fill this gap by providing a state-of-the-art analysis of generative AI in construction, with three objectives: (1) to review and categorize the existing and emerging generative AI opportunities and challenges in the construction industry; (2) to propose a framework for construction firms to build customized generative AI solutions using their own data, comprising steps such as data collection, dataset curation, training custom large language model (LLM), model evaluation, and deployment; and (3) to demonstrate the framework via a case study of developing a generative model for querying contract documents. The results show that retrieval augmented generation (RAG) improves the baseline LLM by 5.2, 9.4, and 4.8% in terms of quality, relevance, and reproducibility. This study provides academics and construction professionals with a comprehensive analysis and practical framework to guide the adoption of generative AI techniques to enhance productivity, quality, safety, and sustainability across the construction industry., Comment: 74 pages, 11 figures, 20 tables

The increasing size of large language models (LLMs) has introduced challenges in their training and inference. Removing model components is perceived as a solution to tackle the large model sizes, however, existing pruning methods solely focus on performance, without considering an essential aspect for the responsible use of LLMs: model fairness. It is crucial to address the fairness of LLMs towards diverse groups, such as women, Black people, LGBTQ+, Jewish communities, among others, as they are being deployed and available to a wide audience. In this work, first, we investigate how attention heads impact fairness and performance in pre-trained transformer-based language models. We then propose a novel method to prune the attention heads that negatively impact fairness while retaining the heads critical for performance, i.e. language modeling capabilities. Our approach is practical in terms of time and resources, as it does not require fine-tuning the final pruned, and fairer, model. Our findings demonstrate a reduction in gender bias by 19%, 19.5%, 39.5%, 34.7%, 23%, and 8% for DistilGPT-2, GPT-2, GPT-Neo of two different sizes, GPT-J, and Llama 2 models, respectively, in comparison to the biased model, with only a slight decrease in performance., Comment: In Proceedings of AAAI 2024