The diversity across outputs generated by large language models shapes the perception of their quality and utility. Prompt leaks, templated answer structure, and canned responses across different interactions are readily noticed by people, but there is no standard score to measure this aspect of model behavior. In this work we empirically investigate diversity scores on English texts. We find that computationally efficient compression algorithms capture information similar to what is measured by slow to compute $n$-gram overlap homogeneity scores. Further, a combination of measures -- compression ratios, self-repetition of long $n$-grams and Self-BLEU and BERTScore -- are sufficient to report, as they have low mutual correlation with each other. The applicability of scores extends beyond analysis of generative models; for example, we highlight applications on instruction-tuning datasets and human-produced texts. We release a diversity score package to facilitate research and invite consistency across reports.

Communication overhead is a significant bottleneck in federated learning (FL), which has been exaggerated with the increasing size of AI models. In this paper, we propose FedRDMA, a communication-efficient cross-silo FL system that integrates RDMA into the FL communication protocol. To overcome the limitations of RDMA in wide-area networks (WANs), FedRDMA divides the updated model into chunks and designs a series of optimization techniques to improve the efficiency and robustness of RDMA-based communication. We implement FedRDMA atop the industrial federated learning framework and evaluate it on a real-world cross-silo FL scenario. The experimental results show that \sys can achieve up to 3.8$\times$ speedup in communication efficiency compared to traditional TCP/IP-based FL systems.

Flaky tests can pass or fail non-deterministically, without alterations to a software system. Such tests are frequently encountered by developers and hinder the credibility of test suites. State-of-the-art research incorporates machine learning solutions into flaky test detection and achieves reasonably good accuracy. Moreover, the majority of automated flaky test repair solutions are designed for specific types of flaky tests. This research work proposes a novel categorization framework, called FlaKat, which uses machine-learning classifiers for fast and accurate prediction of the category of a given flaky test that reflects its root cause. Sampling techniques are applied to address the imbalance between flaky test categories in the International Dataset of Flaky Test (IDoFT). A new evaluation metric, called Flakiness Detection Capacity (FDC), is proposed for measuring the accuracy of classifiers from the perspective of information theory and provides proof for its effectiveness. The final FDC results are also in agreement with F1 score regarding which classifier yields the best flakiness classification.

Geometric graph is a special kind of graph with geometric features, which is vital to model many scientific problems. Unlike generic graphs, geometric graphs often exhibit physical symmetries of translations, rotations, and reflections, making them ineffectively processed by current Graph Neural Networks (GNNs). To tackle this issue, researchers proposed a variety of Geometric Graph Neural Networks equipped with invariant/equivariant properties to better characterize the geometry and topology of geometric graphs. Given the current progress in this field, it is imperative to conduct a comprehensive survey of data structures, models, and applications related to geometric GNNs. In this paper, based on the necessary but concise mathematical preliminaries, we provide a unified view of existing models from the geometric message passing perspective. Additionally, we summarize the applications as well as the related datasets to facilitate later research for methodology development and experimental evaluation. We also discuss the challenges and future potential directions of Geometric GNNs at the end of this survey.

End-to-end (E2E) approach is gradually replacing hybrid models for automatic speech recognition (ASR) tasks. However, the optimization of E2E models lacks an intuitive method for handling decoding shifts, especially in scenarios with a large number of domain-specific rare words that hold specific important meanings. Furthermore, the absence of knowledge-intensive speech datasets in academia has been a significant limiting factor, and the commonly used speech corpora exhibit significant disparities with realistic conversation. To address these challenges, we present Medical Interview (MED-IT), a multi-turn consultation speech dataset that contains a substantial number of knowledge-intensive named entities. We also explore methods to enhance the recognition performance of rare words for E2E models. We propose a novel approach, post-decoder biasing, which constructs a transform probability matrix based on the distribution of training transcriptions. This guides the model to prioritize recognizing words in the biasing list. In our experiments, for subsets of rare words appearing in the training speech between 10 and 20 times, and between 1 and 5 times, the proposed method achieves a relative improvement of 9.3% and 5.1%, respectively.

Large Language Models (LLMs) are prone to factuality hallucination, generating text that contradicts established knowledge. While extensive research has addressed this in English, little is known about multilingual LLMs. This paper systematically evaluates multilingual LLMs' factual accuracy across languages and geographic regions. We introduce a novel pipeline for multilingual factuality evaluation, adapting FActScore(Min et al., 2023) for diverse languages. Our analysis across nine languages reveals that English consistently outperforms others in factual accuracy and quantity of generated facts. Furthermore, multilingual models demonstrate a bias towards factual information from Western continents. These findings highlight the need for improved multilingual factuality assessment and underscore geographical biases in LLMs' fact generation.

Real-world data is often incomplete and contains missing values. To train accurate models over real-world datasets, users need to spend a substantial amount of time and resources imputing and finding proper values for missing data items. In this paper, we demonstrate that it is possible to learn accurate models directly from data with missing values for certain training data and target models. We propose a unified approach for checking the necessity of data imputation to learn accurate models across various widely-used machine learning paradigms. We build efficient algorithms with theoretical guarantees to check this necessity and return accurate models in cases where imputation is unnecessary. Our extensive experiments indicate that our proposed algorithms significantly reduce the amount of time and effort needed for data imputation without imposing considerable computational overhead.

Who is behind the most notorious "deepfake" app on the internet? Trying to answer that question these past few months, for a new Guardian podcast series, Black Box, has been like wandering through a hall of mirrors. The app, ClothOff, has hundreds of thousands of followers and has already been used in a least two cases to generate dozens of images of underage girls – pictures that have left the girls traumatised, their parents outraged and the police baffled at how to stop it. Producers Josh Kelly, Alex Atack and I have followed ClothOff's trail to nondescript addresses in central London that appear to be unoccupied. We have encountered sham businesses, distorted voices and photographs of fake employees.

Hush-hush meetings, often never made public. For centuries, the art of conflict mediation has relied on nuanced human skills: from elements as simple as how to make eye contact and listen carefully to detecting shifts in emotions and subtle signals from opponents. Now, a growing set of entrepreneurs and experts are pitching a dramatic new set of tools into the world of dispute resolution – relying increasingly on artificial intelligence (AI). "Groundbreaking technological advancements are revolutionising the frontier of peace and mediation," said Sama al-Hamdani, programme director of Hala System, a private company using AI and data analysis to gather unencrypted intelligence in conflict zones, among other war-related tasks. "We are witnessing an era where AI transforms mediators into powerhouses of efficiency and insight," al-Hamdani said.

This research addresses the pressing challenge of enhancing processing times and detection capabilities in Unmanned Aerial Vehicle (UAV)/drone imagery for global wildfire detection, despite limited datasets. Proposing a Segmented Neural Network (SegNet) selection approach, we focus on reducing feature maps to boost both time resolution and accuracy significantly advancing processing speeds and accuracy in real-time wildfire detection. This paper contributes to increased processing speeds enabling real-time detection capabilities for wildfire, increased detection accuracy of wildfire, and improved detection capabilities of early wildfire, through proposing a new direction for image classification of amorphous objects like fire, water, smoke, etc. Employing Convolutional Neural Networks (CNNs) for image classification, emphasizing on the reduction of irrelevant features vital for deep learning processes, especially in live feed data for fire detection. Amidst the complexity of live feed data in fire detection, our study emphasizes on image feed, highlighting the urgency to enhance real-time processing. Our proposed algorithm combats feature overload through segmentation, addressing challenges arising from diverse features like objects, colors, and textures. Notably, a delicate balance of feature map size and dataset adequacy is pivotal. Several research papers use smaller image sizes, compromising feature richness which necessitating a new approach. We illuminate the critical role of pixel density in retaining essential details, especially for early wildfire detection. By carefully selecting number of filters during training, we underscore the significance of higher pixel density for proper feature selection. The proposed SegNet approach is rigorously evaluated using real-world dataset obtained by a drone flight and compared to state-of-the-art literature.