This contrasts with traditional models that process text in a single direction, and it has been widely demonstrated that BERT effectively captures contextual and semantic relationships in text, thereby providing a more comprehensive understanding of context. The embedding components of BERT include word embeddings, segment embeddings, and position embeddings. In essence, word embeddings map each word individually into a vector within a high-dimensional space. The segment embeddings allow BERT to differentiate and process single texts or pairs of texts, thereby enabling the understanding of semantic information at the sentence level. The position embeddings provide sequential information to the structure, allowing the model to mark the position of words in a sentence, which aids in further processing at a higher level. Finally, the CLS token at the beginning of the input sequence represents the final hidden state in the embedding vector, which is commonly used as the representation of the entire input sequence.

It's supposed to be the reliable smart assistant that'makes your life easier' with instant titbits of information. But a shocking report has revealed that in many cases, Amazon's Alexa doesn't know the difference between right and wrong. An investigation by Full Fact has found that Alexa spouts incorrect information on topics ranging from MPs' expenses to the origins of the Northern Lights. Full Fact, the UK's independent fact checking organisation, called the findings'misleading' and'clearly a big problem'. What's more, staff at the organization have been furious to discover that Alexa was attributing the wrong answers to none other than Full Fact.

The concept of diversity is widely used in various applications: from image or molecule generation to recommender systems. Thus, being able to properly measure diversity is important. This paper addresses the problem of quantifying diversity for a set of objects. First, we make a systematic review of existing diversity measures and explore their undesirable behavior in some cases. Based on this review, we formulate three desirable properties (axioms) of a reliable diversity measure: monotonicity, uniqueness, and continuity. We show that none of the existing measures has all three properties and thus these measures are not suitable for quantifying diversity. Then, we construct two examples of measures that have all the desirable properties, thus proving that the list of axioms is not self-contradicting. Unfortunately, the constructed examples are too computationally complex for practical use, thus we pose an open problem of constructing a diversity measure that has all the listed properties and can be computed in practice.

We consider the contextual combinatorial bandit setting where in each round, the learning agent, e.g., a recommender system, selects a subset of "arms," e.g., products, and observes rewards for both the individual base arms, which are a function of known features (called "context"), and the super arm (the subset of arms), which is a function of the base arm rewards. The agent's goal is to simultaneously learn the unknown reward functions and choose the highest-reward arms. For example, the "reward" may represent a user's probability of clicking on one of the recommended products. Conventional bandit models, however, employ restrictive reward function models in order to obtain performance guarantees. We make use of deep neural networks to estimate and learn the unknown reward functions and propose Neural UCB Clustering (NeUClust), which adopts a clustering approach to select the super arm in every round by exploiting underlying structure in the context space. Unlike prior neural bandit works, NeUClust uses a neural network to estimate the super arm reward and select the super arm, thus eliminating the need for a known optimization oracle. We non-trivially extend prior neural combinatorial bandit works to prove that NeUClust achieves $\widetilde{O}\left(\widetilde{d}\sqrt{T}\right)$ regret, where $\widetilde{d}$ is the effective dimension of a neural tangent kernel matrix, $T$ the number of rounds. Experiments on real world recommendation datasets show that NeUClust achieves better regret and reward than other contextual combinatorial and neural bandit algorithms.

The upgraded Magnifier app stands out with iOS 18. Technology can be wonderfully convenient and provide a great deal of entertainment, but it can also be a great way to improve your everyday life, too. For those who experience visual challenges, a variety of apps and features can help you. That's why we love this question about apps and features that can help visually challenged loved ones: "I am not tech savvy. I need to know if there is an app that I can download on a phone, that will allow my mother to tell the app, without needing internet services, who she wants to make a phone call to? She's losing her eyesight and can no longer see the numbers on her phone. She's 88 years old and doesn't own a computer and has limited income," writes "Sheryl" of Westminster, Colorado.

Algorithmic intermediaries govern the digital public sphere through their architectures, amplification algorithms, and moderation practices. In doing so, they shape public communication and distribute attention in ways that were previously infeasible with such subtlety, speed and scale. From misinformation and affective polarisation to hate speech and radicalisation, the many pathologies of the digital public sphere attest that they could do so better. But what ideals should they aim at? Political philosophy should be able to help, but existing theories typically assume that a healthy public sphere will spontaneously emerge if only we get the boundaries of free expression right. They offer little guidance on how to intentionally constitute the digital public sphere. In addition to these theories focused on expression, we need a further theory of communicative justice, targeted specifically at the algorithmic intermediaries that shape communication and distribute attention. This lecture argues that political philosophy urgently owes an account of how to govern communication in the digital public sphere, and introduces and defends a democratic egalitarian theory of communicative justice.

Recommenders take place on a wide scale of e-commerce systems, reducing the problem of information overload. The most common approach is to choose a recommender used by the system to make predictions. However, users vary from each other; thus, a one-fits-all approach seems to be sub-optimal. In this paper, we propose a meta-hybrid recommender that uses machine learning to predict an optimal algorithm. In this way, the best-performing recommender is used for each specific session and user. This selection depends on contextual and preferential information collected about the user. We use standard MovieLens and The Movie DB datasets for offline evaluation. We show that based on the proposed model, it is possible to predict which recommender will provide the most precise recommendations to a user. The theoretical performance of our meta-hybrid outperforms separate approaches by 20-50% in normalized Discounted Gain and Root Mean Square Error metrics. However, it is hard to obtain the optimal performance based on widely-used standard information stored about users.

A lover of magazines may find a few good reasons to pay attention to AFM, a new publication about sex and relationships. It's also the latest in a long line of magazines to exist only because of the largesse of a tech company. AFM stands for both "A Fucking Magazine" and "A Feeld Magazine"--that second one a reference to the dating app that is funding the enterprise. Feeld started its life in 2014 specifically to facilitate threesomes. It was originally called 3nder, pronounced "Thrinder," which quickly led the company to receive a trademark-infringement complaint from Tinder.

In recent years, graph neural networks (GNNs) have been commonly utilized for social recommendation systems. However, real-world scenarios often present challenges related to user privacy and business constraints, inhibiting direct access to valuable social information from other platforms. While many existing methods have tackled matrix factorization-based social recommendations without direct social data access, developing GNN-based federated social recommendation models under similar conditions remains largely unexplored. To address this issue, we propose a novel vertical federated social recommendation method leveraging privacy-preserving two-party graph convolution networks (P4GCN) to enhance recommendation accuracy without requiring direct access to sensitive social information. First, we introduce a Sandwich-Encryption module to ensure comprehensive data privacy during the collaborative computing process. Second, we provide a thorough theoretical analysis of the privacy guarantees, considering the participation of both curious and honest parties. Extensive experiments on four real-world datasets demonstrate that P4GCN outperforms state-of-the-art methods in terms of recommendation accuracy. The code is available at https://github.com/WwZzz/P4GCN.

When annotators disagree, predicting the labels given by individual annotators can capture nuances overlooked by traditional label aggregation. We introduce three approaches to predicting individual annotator ratings on the toxicity of text by incorporating individual annotator-specific information: a neural collaborative filtering (NCF) approach, an in-context learning (ICL) approach, and an intermediate embedding-based architecture. We also study the utility of demographic information for rating prediction. NCF showed limited utility; however, integrating annotator history, demographics, and survey information permits both the embedding-based architecture and ICL to substantially improve prediction accuracy, with the embedding-based architecture outperforming the other methods. We also find that, if demographics are predicted from survey information, using these imputed demographics as features performs comparably to using true demographic data. This suggests that demographics may not provide substantial information for modeling ratings beyond what is captured in survey responses. Our findings raise considerations about the relative utility of different types of annotator information and provide new approaches for modeling annotators in subjective NLP tasks.