Whisper, the recently developed multilingual weakly supervised model, is reported to perform well on multiple speech recognition benchmarks in both monolingual and multilingual settings. However, it is not clear how Whisper would fare under diverse conditions even on languages it was evaluated on such as Arabic. In this work, we address this gap by comprehensively evaluating Whisper on several varieties of Arabic speech for the ASR task. Our evaluation covers most publicly available Arabic speech data and is performed under n-shot (zero-, few-, and full) finetuning. We also investigate the robustness of Whisper under completely novel conditions, such as in dialect-accented standard Arabic and in unseen dialects for which we develop evaluation data. Our experiments show that although Whisper zero-shot outperforms fully finetuned XLS-R models on all datasets, its performance deteriorates significantly in the zero-shot setting for five unseen dialects (i.e., Algeria, Jordan, Palestine, UAE, and Yemen).

This study aimed to develop an algorithm to automatically extract annotations for chest X-ray classification models from German thoracic radiology reports. An automatic label extraction model was designed based on the CheXpert architecture, and a web-based annotation interface was created for iterative improvements. Results showed that automated label extraction can reduce time spent on manual labeling and improve overall modeling performance. The model trained on automatically extracted labels performed competitively to manually labeled data and strongly outperformed the model trained on publicly available data.

Reasoning about knowledge among multiple agents plays an important role in studying real-world problems in a distributed setting, e.g., in communicating processes, protocols, strategies and games. Multi-agent epistemic logic (EL) [1] and its dynamic extensions, popularly known as dynamic epistemic logics (DEL) [2] are well-known logical systems to specify and reason about such dynamic interactions of knowledge. Traditionally, agents' knowledge is about facts and EL/DEL mostly deals with this phenomenon of'knowing that'. More recently, the notions of'knowing whether', 'knowing why' and'knowing how' have also been investigated from a formal viewpoint [3]. These agents also have expectations about the world around them, and they reason based on what they observe around them, and such observations may or may not match the expectations they have about their surroundings.

Adapters have emerged as a modular and parameter-efficient approach to (zero-shot) cross-lingual transfer. The established MAD-X framework employs separate language and task adapters which can be arbitrarily combined to perform the transfer of any task to any target language. Subsequently, BAD-X, an extension of the MAD-X framework, achieves improved transfer at the cost of MAD-X's modularity by creating "bilingual" adapters specific to the source-target language pair. In this work, we aim to take the best of both worlds by (i) fine-tuning task adapters adapted to the target language(s) (so-called "target language-ready" (TLR) adapters) to maintain high transfer performance, but (ii) without sacrificing the highly modular design of MAD-X. The main idea of "target language-ready" adapters is to resolve the training-vs-inference discrepancy of MAD-X: the task adapter "sees" the target language adapter for the very first time during inference, and thus might not be fully compatible with it. We address this mismatch by exposing the task adapter to the target language adapter during training, and empirically validate several variants of the idea: in the simplest form, we alternate between using the source and target language adapters during task adapter training, which can be generalized to cycling over any set of language adapters. We evaluate different TLR-based transfer configurations with varying degrees of generality across a suite of standard cross-lingual benchmarks, and find that the most general (and thus most modular) configuration consistently outperforms MAD-X and BAD-X on most tasks and languages.

Colexification refers to the linguistic phenomenon where a single lexical form is used to convey multiple meanings. By studying cross-lingual colexifications, researchers have gained valuable insights into fields such as psycholinguistics and cognitive sciences [Jackson et al.,2019]. While several multilingual colexification datasets exist, there is untapped potential in using this information to bootstrap datasets across such semantic features. In this paper, we aim to demonstrate how colexifications can be leveraged to create such cross-lingual datasets. We showcase curation procedures which result in a dataset covering 142 languages across 21 language families across the world. The dataset includes ratings of concreteness and affectiveness, mapped with phonemes and phonological features. We further analyze the dataset along different dimensions to demonstrate potential of the proposed procedures in facilitating further interdisciplinary research in psychology, cognitive science, and multilingual natural language processing (NLP). Based on initial investigations, we observe that i) colexifications that are closer in concreteness/affectiveness are more likely to colexify; ii) certain initial/last phonemes are significantly correlated with concreteness/affectiveness intra language families, such as /k/ as the initial phoneme in both Turkic and Tai-Kadai correlated with concreteness, and /p/ in Dravidian and Sino-Tibetan correlated with Valence; iii) the type-to-token ratio (TTR) of phonemes are positively correlated with concreteness across several language families, while the length of phoneme segments are negatively correlated with concreteness; iv) certain phonological features are negatively correlated with concreteness across languages. The dataset is made public online for further research.

Adversarial examples bring a considerable security threat to support vector machines (SVMs), especially those used in safety-critical applications. Thus, robustness verification is an essential issue for SVMs, which can provide provable robustness against various kinds of adversary attacks. The evaluation results obtained through the robustness verification can provide a safe guarantee for the use of SVMs. The existing verification method does not often perform well in verifying SVMs with nonlinear kernels. To this end, we propose a method to improve the verification performance for SVMs with nonlinear kernels. We first formalize the adversarial robustness evaluation of SVMs as an optimization problem. Then a lower bound of the original problem is obtained by solving the Lagrangian dual problem of the original problem. Finally, the adversarial robustness of SVMs is evaluated concerning the lower bound. We evaluate the adversarial robustness of SVMs with linear and nonlinear kernels on the MNIST and Fashion-MNIST datasets. The experimental results show that the percentage of provable robustness obtained by our method on the test set is better than that of the state-of-the-art.

This project demonstrates how medical corpus hypothesis generation, a knowledge discovery field of AI, can be used to derive new research angles for landscape and urban planners. The hypothesis generation approach herein consists of a combination of deep learning with topic modeling, a probabilistic approach to natural language analysis that scans aggregated research databases for words that can be grouped together based on their subject matter commonalities; the word groups accordingly form topics that can provide implicit connections between two general research terms. The hypothesis generation system AGATHA was used to identify likely conceptual relationships between emerging infectious diseases (EIDs) and deforestation, with the objective of providing landscape planners guidelines for productive research directions to help them formulate research hypotheses centered on deforestation and EIDs that will contribute to the broader health field that asserts causal roles of landscape-level issues. This research also serves as a partial proof-of-concept for the application of medical database hypothesis generation to medicine-adjacent hypothesis discovery. Keywords deforestation, emerging infectious disease, hypothesis generation, landscape planning, topic modeling Funding This research was funded by National Science Foundation grant numbers 1633608 and 2027864. The authors express their gratitude to the NSF for its generous support of their research. Introduction The recent COVID-19 crisis has put the issue of emerging infectious diseases (EIDs) back in the global spotlight. Addressing EIDs going forward will require widespread interdisciplinary cooperation, as discouraging them is a multifaceted and omnipresent endeavor. Biologists and health experts have frequently asserted that landscape-level issues drive EIDs (e.g.

Healthcare in Africa is a complex issue influenced by many factors including poverty, lack of infrastructure, and inadequate funding. However, Artificial intelligence (AI) applied to healthcare, has the potential to transform healthcare in Africa by improving the accuracy and efficiency of diagnosis, enabling earlier detection of diseases, and supporting the delivery of personalized medicine. This paper reviews the current state of how AI Algorithms can be used to improve diagnostics, treatment, and disease monitoring, as well as how AI can be used to improve access to healthcare in Africa as a low-resource setting and discusses some of the critical challenges and opportunities for its adoption. As such, there is a need for a well-coordinated effort by the governments, private sector, healthcare providers, and international organizations to create sustainable AI solutions that meet the unique needs of the African healthcare system.

Composed image retrieval aims to find an image that best matches a given multi-modal user query consisting of a reference image and text pair. Existing methods commonly pre-compute image embeddings over the entire corpus and compare these to a reference image embedding modified by the query text at test time. Such a pipeline is very efficient at test time since fast vector distances can be used to evaluate candidates, but modifying the reference image embedding guided only by a short textual description can be difficult, especially independent of potential candidates. An alternative approach is to allow interactions between the query and every possible candidate, i.e., reference-text-candidate triplets, and pick the best from the entire set. Though this approach is more discriminative, for large-scale datasets the computational cost is prohibitive since pre-computation of candidate embeddings is no longer possible. We propose to combine the merits of both schemes using a two-stage model. Our first stage adopts the conventional vector distancing metric and performs a fast pruning among candidates. Meanwhile, our second stage employs a dual-encoder architecture, which effectively attends to the input triplet of reference-text-candidate and re-ranks the candidates. Both stages utilize a vision-and-language pre-trained network, which has proven beneficial for various downstream tasks. Our method consistently outperforms state-of-the-art approaches on standard benchmarks for the task.

Layer-wise distillation is a powerful tool to compress large models (i.e. teacher models) into small ones (i.e., student models). The student distills knowledge from the teacher by mimicking the hidden representations of the teacher at every intermediate layer. However, layer-wise distillation is difficult. Since the student has a smaller model capacity than the teacher, it is often under-fitted. Furthermore, the hidden representations of the teacher contain redundant information that the student does not necessarily need for the target task's learning. To address these challenges, we propose a novel Task-aware layEr-wise Distillation (TED). TED designs task-aware filters to align the hidden representations of the student and the teacher at each layer. The filters select the knowledge that is useful for the target task from the hidden representations. As such, TED reduces the knowledge gap between the two models and helps the student to fit better on the target task. We evaluate TED in two scenarios: continual pre-training and fine-tuning. TED demonstrates significant and consistent improvements over existing distillation methods in both scenarios. Code is available at https://github.com/cliang1453/task-aware-distillation.