Search

Job interviews play a critical role in shaping one's career, yet practicing interview skills can be challenging, especially without access to human coaches or peers for feedback. Recent advancements in large language models (LLMs) present an opportunity to enhance the interview practice experience. Yet, little research has explored the effectiveness and user perceptions of such systems or the benefits and challenges of using LLMs for interview practice. Furthermore, while prior work and recent commercial tools have demonstrated the potential of AI to assist with interview practice, they often deliver one-way feedback, where users only receive information about their performance. By contrast, dialogic feedback, a concept developed in learning sciences, is a two-way interaction feedback process that allows users to further engage with and learn from the provided feedback through interactive dialogue. This paper introduces Conversate, a web-based application that supports reflective learning in job interview practice by leveraging large language models (LLMs) for interactive interview simulations and dialogic feedback. To start the interview session, the user provides the title of a job position (e.g., entry-level software engineer) in the system. Then, our system will initialize the LLM agent to start the interview simulation by asking the user an opening interview question and following up with questions carefully adapted to subsequent user responses. After the interview session, our back-end LLM framework will then analyze the user's responses and highlight areas for improvement. Users can then annotate the transcript by selecting specific sections and writing self-reflections. Finally, the user can interact with the system for dialogic feedback, conversing with the LLM agent to learn from and iteratively refine their answers based on the agent's guidance.

Online hate speech has become increasingly prevalent on social media platforms, causing harm to individuals and society. While efforts have been made to combat this issue through content moderation, the potential of user-driven counterspeech as an alternative solution remains underexplored. Existing counterspeech methods often face challenges such as fear of retaliation and skill-related barriers. To address these challenges, we introduce CounterQuill, an AI-mediated system that assists users in composing effective and empathetic counterspeech. CounterQuill provides a three-step process: (1) a learning session to help users understand hate speech and counterspeech; (2) a brainstorming session that guides users in identifying key elements of hate speech and exploring counterspeech strategies; and (3) a co-writing session that enables users to draft and refine their counterspeech with CounterQuill. We conducted a within-subjects user study with 20 participants to evaluate CounterQuill in comparison to ChatGPT. Results show that CounterQuill's guidance and collaborative writing process provided users a stronger sense of ownership over their co-authored counterspeech. Users perceived CounterQuill as a writing partner and thus were more willing to post the co-written counterspeech online compared to the one written with ChatGPT.

Cardiac auscultation, an integral tool in diagnosing cardiovascular diseases (CVDs), often relies on the subjective interpretation of clinicians, presenting a limitation in consistency and accuracy. Addressing this, we introduce the BUET Multi-disease Heart Sound (BMD-HS) dataset - a comprehensive and meticulously curated collection of heart sound recordings. This dataset, encompassing 864 recordings across five distinct classes of common heart sounds, represents a broad spectrum of valvular heart diseases, with a focus on diagnostically challenging cases. The standout feature of the BMD-HS dataset is its innovative multi-label annotation system, which captures a diverse range of diseases and unique disease states. This system significantly enhances the dataset's utility for developing advanced machine learning models in automated heart sound classification and diagnosis. By bridging the gap between traditional auscultation practices and contemporary data-driven diagnostic methods, the BMD-HS dataset is poised to revolutionize CVD diagnosis and management, providing an invaluable resource for the advancement of cardiac health research. The dataset is publicly available at this link: https://github.com/mHealthBuet/BMD-HS-Dataset., Comment: 14 pages, 13 figures

Studying the dynamics of quantum many-body systems is often constrained by the limitations in probing relevant observables, especially in continuous systems. A powerful method to gain information about such systems is the reconstruction of local currents from the continuity equation. Here we extend this approach to extract the total phase fluctuation of adjacent Bose gases. We validate our technique numerically and demonstrate its effectiveness by analyzing data from selected experiments simulating 1D quantum field theories through the phase difference of two 1D Bose gases probed by interference. Our analysis reveals the previously hidden sector of the sum mode of the phase, which is important for studying long-time thermalization and out-of-equilibrium dynamics of the system, thereby expanding the scope and capabilities of cold-atomic quantum simulators., Comment: 5+9 pages, 12 figures

News media structure their reporting of events or issues using certain perspectives. When describing an incident involving gun violence, for example, some journalists may focus on mental health or gun regulation, while others may emphasize the discussion of gun rights. Such perspectives are called \say{frames} in communication research. We study, for the first time, the value of combining lead images and their contextual information with text to identify the frame of a given news article. We observe that using multiple modes of information(article- and image-derived features) improves prediction of news frames over any single mode of information when the images are relevant to the frames of the headlines. We also observe that frame image relevance is related to the ease of conveying frames via images, which we call frame concreteness. Additionally, we release the first multimodal news framing dataset related to gun violence in the U.S., curated and annotated by communication researchers. The dataset will allow researchers to further examine the use of multiple information modalities for studying media framing., Comment: published at Findings of the Association for Computational Linguistics: EMNLP 2021

Despite the advancement of deep learning-based computer-aided diagnosis (CAD) methods for pneumonia from adult chest x-ray (CXR) images, the performance of CAD methods applied to pediatric images remains suboptimal, mainly due to the lack of large-scale annotated pediatric imaging datasets. Establishing a proper framework to leverage existing adult large-scale CXR datasets can thus enhance pediatric pneumonia detection performance. In this paper, we propose a three-branch parallel path learning-based framework that utilizes both adult and pediatric datasets to improve the performance of deep learning models on pediatric test datasets. The paths are trained with pediatric only, adult only, and both types of CXRs, respectively. Our proposed framework utilizes the multi-positive contrastive loss to cluster the classwise embeddings and the embedding similarity loss among these three parallel paths to make the classwise embeddings as close as possible to reduce the effect of domain shift. Experimental evaluations on open-access adult and pediatric CXR datasets show that the proposed method achieves a superior AUROC score of 0.8464 compared to 0.8348 obtained using the conventional approach of join training on both datasets. The proposed approach thus paves the way for generalized CAD models that are effective for both adult and pediatric age groups., Comment: Accepted to International Conference of IEEE Engineering in Medicine and Biology Society (EMBC), 2024

Objective: Lung auscultation is a valuable tool in diagnosing and monitoring various respiratory diseases. However, lung sounds (LS) are significantly affected by numerous sources of contamination, especially when recorded in real-world clinical settings. Conventional denoising models prove impractical for LS denoising, primarily owing to spectral overlap complexities arising from diverse noise sources. To address this issue, we propose a specialized deep-learning model (Uformer) for lung sound denoising. Methods: The proposed Uformer model is constituted of three modules: a Convolutional Neural Network (CNN) encoder module, dedicated to extracting latent features; a Transformer encoder module, employed to further enhance the encoding of unique LS features and effectively capture intricate long-range dependencies; and a CNN decoder module, employed to generate the denoised signals. An ablation study was performed in order to find the most optimal architecture. Results: The performance of the proposed Uformer model was evaluated on lung sounds induced with different types of synthetic and real-world noises. Lung sound signals of -12 dB to 15 dB signal-to-noise ratio (SNR) were considered in testing experiments. The proposed model showed an average SNR improvement of 16.51 dB when evaluated with -12 dB LS signals. Our end-to-end model, with an average SNR improvement of 19.31 dB, outperforms the existing model when evaluated with ambient noise and fewer parameters. Conclusion: Based on the qualitative and quantitative findings in this study, it can be stated that Uformer is robust and generalized to be used in assisting the monitoring of respiratory conditions.

Spatially resolved relative phase measurement of two adjacent 1D Bose gases is enabled by matter-wave interference upon free expansion. However, longitudinal dynamics is typically ignored in the analysis of experimental data. We provide an analytical formula showing a correction to the readout of the relative phase due to longitudinal expansion and mixing with the common phase. We numerically assess the error propagation to the estimation of the gases' physical quantities such as correlation functions and temperature. Our work characterizes the reliability and robustness of interferometric measurements, directing us to the improvement of existing phase extraction methods necessary to observe new physical phenomena in cold-atomic quantum simulators., Comment: Significant updates from v1, contains new results section on the effect of image processing, 21+12 pages, 22 figures

In machine learning fairness, training models that minimize disparity across different sensitive groups often leads to diminished accuracy, a phenomenon known as the fairness-accuracy trade-off. The severity of this trade-off inherently depends on dataset characteristics such as dataset imbalances or biases and therefore, using a uniform fairness requirement across diverse datasets remains questionable. To address this, we present a computationally efficient approach to approximate the fairness-accuracy trade-off curve tailored to individual datasets, backed by rigorous statistical guarantees. By utilizing the You-Only-Train-Once (YOTO) framework, our approach mitigates the computational burden of having to train multiple models when approximating the trade-off curve. Crucially, we introduce a novel methodology for quantifying uncertainty in our estimates, thereby providing practitioners with a robust framework for auditing model fairness while avoiding false conclusions due to estimation errors. Our experiments spanning tabular (e.g., Adult), image (CelebA), and language (Jigsaw) datasets underscore that our approach not only reliably quantifies the optimum achievable trade-offs across various data modalities but also helps detect suboptimality in SOTA fairness methods.

Off-Policy Evaluation (OPE) in contextual bandits is crucial for assessing new policies using existing data without costly experimentation. However, current OPE methods, such as Inverse Probability Weighting (IPW) and Doubly Robust (DR) estimators, suffer from high variance, particularly in cases of low overlap between target and behavior policies or large action and context spaces. In this paper, we introduce a new OPE estimator for contextual bandits, the Marginal Ratio (MR) estimator, which focuses on the shift in the marginal distribution of outcomes $Y$ instead of the policies themselves. Through rigorous theoretical analysis, we demonstrate the benefits of the MR estimator compared to conventional methods like IPW and DR in terms of variance reduction. Additionally, we establish a connection between the MR estimator and the state-of-the-art Marginalized Inverse Propensity Score (MIPS) estimator, proving that MR achieves lower variance among a generalized family of MIPS estimators. We further illustrate the utility of the MR estimator in causal inference settings, where it exhibits enhanced performance in estimating Average Treatment Effects (ATE). Our experiments on synthetic and real-world datasets corroborate our theoretical findings and highlight the practical advantages of the MR estimator in OPE for contextual bandits., Comment: Conference on Neural Information Processing Systems (NeurIPS 2023)

We intend to study a modified version of the planar Circular Restricted Three-Body Problem (CRTBP) by incorporating several perturbing parameters. We consider the bigger primary as an oblate spheroid and emitting radiation while the small primary has an elongated body. We also consider the perturbation from a disk-like structure encompassing this three-body system. First, we develop a mathematical model of this modified CRTBP. We have found there exist five equilibrium points in this modified CRTBP model, where three of them are collinear and the other two are non-collinear. Second, we apply our modified CRTBP model to the Sun-Haumea system by considering several values of each perturbing parameter. Through our numerical investigation, we have discovered that the incorporation of perturbing parameters has resulted in a shift in the equilibrium point positions of the Sun-Haumea system compared to their positions in the classical CRTBP. The stability of equilibrium points is investigated. We have shown that the collinear equilibrium points are unstable and the stability of non-collinear equilibrium points depends on the mass parameter $\mu$ of the system. Unlike the classical case, non-collinear equilibrium points have both a maximum and minimum limit of $\mu$ for achieving stability. We remark that the stability range of $\mu$ in non-collinear equilibrium points depends on the perturbing parameters. In context of the Sun-Haumea system, we have found that the non-collinear equilibrium points are stable., Comment: Accepted for publication in Research in Astronomy and Astrophysics

Ensuring alignment, which refers to making models behave in accordance with human intentions [1,2], has become a critical task before deploying large language models (LLMs) in real-world applications. For instance, OpenAI devoted six months to iteratively aligning GPT-4 before its release [3]. However, a major challenge faced by practitioners is the lack of clear guidance on evaluating whether LLM outputs align with social norms, values, and regulations. This obstacle hinders systematic iteration and deployment of LLMs. To address this issue, this paper presents a comprehensive survey of key dimensions that are crucial to consider when assessing LLM trustworthiness. The survey covers seven major categories of LLM trustworthiness: reliability, safety, fairness, resistance to misuse, explainability and reasoning, adherence to social norms, and robustness. Each major category is further divided into several sub-categories, resulting in a total of 29 sub-categories. Additionally, a subset of 8 sub-categories is selected for further investigation, where corresponding measurement studies are designed and conducted on several widely-used LLMs. The measurement results indicate that, in general, more aligned models tend to perform better in terms of overall trustworthiness. However, the effectiveness of alignment varies across the different trustworthiness categories considered. This highlights the importance of conducting more fine-grained analyses, testing, and making continuous improvements on LLM alignment. By shedding light on these key dimensions of LLM trustworthiness, this paper aims to provide valuable insights and guidance to practitioners in the field. Understanding and addressing these concerns will be crucial in achieving reliable and ethically sound deployment of LLMs in various applications., Comment: Fixed several typos

The recent 'Mpox' outbreak, formerly known as 'Monkeypox', has become a significant public health concern and has spread to over 110 countries globally. The challenge of clinically diagnosing mpox early on is due, in part, to its similarity to other types of rashes. Computer-aided screening tools have been proven valuable in cases where Polymerase Chain Reaction (PCR) based diagnosis is not immediately available. Deep learning methods are powerful in learning complex data representations, but their efficacy largely depends on adequate training data. To address this challenge, we present the "Mpox Skin Lesion Dataset Version 2.0 (MSLD v2.0)" as a follow-up to the previously released openly accessible dataset, one of the first datasets containing mpox lesion images. This dataset contains images of patients with mpox and five other non-mpox classes (chickenpox, measles, hand-foot-mouth disease, cowpox, and healthy). We benchmark the performance of several state-of-the-art deep learning models, including VGG16, ResNet50, DenseNet121, MobileNetV2, EfficientNetB3, InceptionV3, and Xception, to classify mpox and other infectious skin diseases. In order to reduce the impact of racial bias, we utilize a color space data augmentation method to increase skin color variability during training. Additionally, by leveraging transfer learning implemented with pre-trained weights generated from the HAM10000 dataset, an extensive collection of pigmented skin lesion images, we achieved the best overall accuracy of $83.59\pm2.11\%$. Finally, the developed models are incorporated within a prototype web application to analyze uploaded skin images by a user and determine whether a subject is a suspected mpox patient.