As the field of machine learning (ML) continues to evolve and its impact on society and various aspects of our lives grows, it is becoming increasingly important for practitioners and innovators to consider a broader range of perspectives when building ML models and applications. This desire is driving the need for a more flexible and scalable ML infrastructure. At Spotify, we strongly believe in a diverse and collaborative approach to building ML applications. Gone are the days when ML was the domain of only a small group of researchers and engineers. We want to democratize our ML efforts such that contributors of all backgrounds, including engineers, data scientists, and researchers, can leverage their unique perspectives, skills, and expertise to further ML at Spotify.

Article: The CEO of one of the world's biggest media companies just said A.I. is making some journalists obsolete as he plans staff cuts - Fortune ChatGPT has gotten people to make all kinds of predictions. One of the main ones is that the Artificial-Intelligence-Chat tool will eliminate the jobs of writers. Whether that is true or not remains to be seen. And, to what degree that might happen also remains to be seen. Regardless, the CEO of German media group Axel Springer, which owns German newspapers Bild and Welt as well as U.S. website Politico says he will be making large cuts because of the AI chat tool.

Pretrained language models (LMs) are susceptible to generate text with nonfactual information. In this work, we measure and improve the factual accuracy of large-scale LMs for open-ended text generation. We design the FactualityPrompts test set and metrics to measure the factuality of LM generations. Based on that, we study the factual accuracy of LMs with parameter sizes ranging from 126M to 530B. Interestingly, we find that larger LMs are more factual than smaller ones, although a previous study suggests that larger LMs can be less truthful in terms of misconceptions. In addition, popular sampling algorithms (e.g., top-p) in open-ended text generation can harm the factuality due to the ''uniform randomness'' introduced at every sampling step. We propose the factual-nucleus sampling algorithm that dynamically adapts the randomness to improve the factuality of generation while maintaining quality. Furthermore, we analyze the inefficiencies of the standard training method in learning correct associations between entities from factual text corpus (e.g., Wikipedia). We propose a factuality-enhanced training method that uses TopicPrefix for better awareness of facts and sentence completion as the training objective, which can vastly reduce the factual errors. We release our code and FactualityPrompts benchmark at: https://github.com/nayeon7lee/FactualityPrompt.

Research and education in machine learning needs diverse, representative, and open datasets that contain sufficient samples to handle the necessary training, validation, and testing tasks. Currently, the Recommender Systems area includes a large number of subfields in which accuracy and beyond accuracy quality measures are continuously improved. To feed this research variety, it is necessary and convenient to reinforce the existing datasets with synthetic ones. This paper proposes a Generative Adversarial Network (GAN)-based method to generate collaborative filtering datasets in a parameterized way, by selecting their preferred number of users, items, samples, and stochastic variability. This parameterization cannot be made using regular GANs. Our GAN model is fed with dense, short, and continuous embedding representations of items and users, instead of sparse, large, and discrete vectors, to make an accurate and quick learning, compared to the traditional approach based on large and sparse input vectors. The proposed architecture includes a DeepMF model to extract the dense user and item embeddings, as well as a clustering process to convert from the dense GAN generated samples to the discrete and sparse ones, necessary to create each required synthetic dataset. The results of three different source datasets show adequate distributions and expected quality values and evolutions on the generated datasets compared to the source ones. Synthetic datasets and source codes are available to researchers.

Modern text classification systems have impressive capabilities but are infeasible to deploy and use reliably due to their dependence on prompting and billion-parameter language models. SetFit (Tunstall et al., 2022) is a recent, practical approach that fine-tunes a Sentence Transformer under a contrastive learning paradigm and achieves similar results to more unwieldy systems. Text classification is important for addressing the problem of domain drift in detecting harmful content, which plagues all social media platforms. Here, we propose Like a Good Nearest Neighbor (LaGoNN), an inexpensive modification to SetFit that requires no additional parameters or hyperparameters but modifies input with information about its nearest neighbor, for example, the label and text, in the training data, making novel data appear similar to an instance on which the model was optimized. LaGoNN is effective at the task of detecting harmful content and generally improves performance compared to SetFit. To demonstrate the value of our system, we conduct a thorough study of text classification systems in the context of content moderation under four label distributions.

This letter describes a method for autonomous aerial cinematography with distributed lighting by a team of unmanned aerial vehicles (UAVs). Although camera-carrying multi-rotor helicopters have become commonplace in cinematography, their usage is limited to scenarios with sufficient natural light or of lighting provided by static artificial lights. We propose to use a formation of unmanned aerial vehicles as a tool for filming a target under illumination from various directions, which is one of the fundamental techniques of traditional cinematography. We decompose the multi-UAV trajectory optimization problem to tackle non-linear cinematographic aspects and obstacle avoidance at separate stages, which allows us to re-plan in real time and react to changes in dynamic environments. The performance of our method has been evaluated in realistic simulation scenarios and field experiments, where we show how it increases the quality of the shots and that it is capable of planning safe trajectories even in cluttered environments.

With the expansion of social media and the increasing dissemination of multimedia content, the spread of misinformation has become a major concern. This necessitates effective strategies for multimodal misinformation detection (MMD) that detect whether the combination of an image and its accompanying text could mislead or misinform. Due to the data-intensive nature of deep neural networks and the labor-intensive process of manual annotation, researchers have been exploring various methods for automatically generating synthetic multimodal misinformation - which we refer to as Synthetic Misinformers - in order to train MMD models. However, limited evaluation on real-world misinformation and a lack of comparisons with other Synthetic Misinformers makes difficult to assess progress in the field. To address this, we perform a comparative study on existing and new Synthetic Misinformers that involves (1) out-of-context (OOC) image-caption pairs, (2) cross-modal named entity inconsistency (NEI) as well as (3) hybrid approaches and we evaluate them against real-world misinformation; using the COSMOS benchmark. The comparative study showed that our proposed CLIP-based Named Entity Swapping can lead to MMD models that surpass other OOC and NEI Misinformers in terms of multimodal accuracy and that hybrid approaches can lead to even higher detection accuracy. Nevertheless, after alleviating information leakage from the COSMOS evaluation protocol, low Sensitivity scores indicate that the task is significantly more challenging than previous studies suggested. Finally, our findings showed that NEI-based Synthetic Misinformers tend to suffer from a unimodal bias, where text-only MMDs can outperform multimodal ones.

This paper describes our approach to the multi-modal fact verification (FACTIFY) challenge at AAAI2023. In recent years, with the widespread use of social media, fake news can spread rapidly and negatively impact social security. Automatic claim verification becomes more and more crucial to combat fake news. In fact verification involving multiple modal data, there should be a structural coherence between claim and document. Therefore, we proposed a structure coherence-based multi-modal fact verification scheme to classify fake news. Our structure coherence includes the following four aspects: sentence length, vocabulary similarity, semantic similarity, and image similarity. Specifically, CLIP and Sentence BERT are combined to extract text features, and ResNet50 is used to extract image features. In addition, we also extract the length of the text as well as the lexical similarity. Then the features were concatenated and passed through the random forest classifier. Finally, our weighted average F1 score has reached 0.8079, achieving 2nd place in FACTIFY2.

A Reflectance Transformation Imaging technique (RTI) realized by multi-rotor Unmanned Aerial Vehicles (UAVs) with a focus on deployment in difficult to access buildings is presented in this letter. RTI is a computational photographic method that captures a surface shape and color of a subject and enables its interactive re-lighting from any direction in a software viewer, revealing details that are not visible with the naked eye. The input of RTI is a set of images captured by a static camera, each one under illumination from a different known direction. We present an innovative approach applying two multi-rotor UAVs to perform this scanning procedure in locations that are hardly accessible or even inaccessible for people. The proposed system is designed for its safe deployment within real-world scenarios in historical buildings with priceless historical value.

Recently, ChatGPT has attracted great attention, as it can generate fluent and high-quality responses to human inquiries. Several prior studies have shown that ChatGPT attains remarkable generation ability compared with existing models. However, the quantitative analysis of ChatGPT's understanding ability has been given little attention. In this report, we explore the understanding ability of ChatGPT by evaluating it on the most popular GLUE benchmark, and comparing it with 4 representative fine-tuned BERT-style models. We find that: 1) ChatGPT falls short in handling paraphrase and similarity tasks; 2) ChatGPT outperforms all BERT models on inference tasks by a large margin; 3) ChatGPT achieves comparable performance compared with BERT on sentiment analysis and question-answering tasks. Additionally, by combining some advanced prompting strategies, we show that the understanding ability of ChatGPT can be further improved.