Given a user's input text, text-matching recommender systems output relevant items by comparing the input text to available items' description, such as product-to-product recommendation on e-commerce platforms. As users' interests and item inventory are expected to change, it is important for a text-matching system to generalize to data shifts, a task known as out-of-distribution (OOD) generalization. However, we find that the popular approach of fine-tuning a large, base language model on paired item relevance data (e.g., user clicks) can be counter-productive for OOD generalization. For a product recommendation task, fine-tuning obtains worse accuracy than the base model when recommending items in a new category or for a future time period. To explain this generalization failure, we consider an intervention-based importance metric, which shows that a fine-tuned model captures spurious correlations and fails to learn the causal features that determine the relevance between any two text inputs. Moreover, standard methods for causal regularization do not apply in this setting, because unlike in images, there exist no universally spurious features in a text-matching task (the same token may be spurious or causal depending on the text it is being matched to). For OOD generalization on text inputs, therefore, we highlight a different goal: avoiding high importance scores for certain features. We do so using an intervention-based regularizer that constraints the causal effect of any token on the model's relevance score to be similar to the base model. Results on Amazon product and 3 question recommendation datasets show that our proposed regularizer improves generalization for both in-distribution and OOD evaluation, especially in difficult scenarios when the base model is not accurate.

Amazon Web Services has been hit with a worldwide outage, impacting dozens of websites that use the company's cloud hosting service. DownDetector, which monitors online outages, shows hundreds of thousands of issue reports from around the globe. Amazon Web Services began experiencing problems around 2:56 pm ET, taking out other websites like IMDB, McDonald's and OkCupid. The e-commerce giant's purchasing platform, music and virtual assistant Alexa are also experiencing problems. Amazon Web Services has been hit with a worldwide outage impacting hundreds of websites that use the company's cloud-hosting service Reports first indicated Amazon Web Services (AWS) was experiencing issues, but other websites began to follow one by one.

Smart plugs are among the simpler smart home devices, giving you voice and app control over appliances like lamps, fans, humidifiers and basic coffee makers. You can create schedules and routines, too, either through a plug's proprietary app or through your preferred smart home platform. But much like other IoT devices, which system plays nice with which plug depends on compatibility, and each brand's app offers different features. We tested out ten popular options to see which are worth buying. Before you buy one, it helps to know what a smart plug can and can't do. They work best with things that have an on/off switch, making them great for lamps and other lights.

Amazon's Echo Studio usually sells for $200 and rarely goes on sale, but you can get the supersize Alexa hub right now in Charcoal or Glacier for $160. We called the Echo Studio the "best sounding speaker Amazon has built" in our review when it was released in 2019. The company has released more speakers and smart displays since then, but the Echo Studio remains a solid option for immersive, high-resolution music streaming. Amazon's Echo Studio speaker is on sale for $160. We gave the Echo Studio a score of 88 in our review and praised it for having excellent audio quality best demonstrated by high-definition and Ultra HD tracks.

Recommender systems have become essential tools for enhancing user experiences across various domains. While extensive research has been conducted on recommender systems for movies, music, and e-commerce, the rapidly growing and economically significant Non-Fungible Token (NFT) market remains underexplored. The unique characteristics and increasing prominence of the NFT market highlight the importance of developing tailored recommender systems to cater to its specific needs and unlock its full potential. In this paper, we examine the distinctive characteristics of NFTs and propose the first recommender system specifically designed to address NFT market challenges. In specific, we develop a Multi-Attention Recommender System for NFTs (NFT-MARS) with three key characteristics: (1) graph attention to handle sparse user-item interactions, (2) multi-modal attention to incorporate feature preference of users, and (3) multi-task learning to consider the dual nature of NFTs as both artwork and financial assets. We demonstrate the effectiveness of NFT-MARS compared to various baseline models using the actual transaction data of NFTs collected directly from blockchain for four of the most popular NFT collections. The source code and data are available at https://anonymous.4open.science/r/RecSys2023-93ED.

Training good representations for items is critical in recommender models. Typically, an item is assigned a unique randomly generated ID, and is commonly represented by learning an embedding corresponding to the value of the random ID. Although widely used, this approach have limitations when the number of items are large and items are power-law distributed -- typical characteristics of real-world recommendation systems. This leads to the item cold-start problem, where the model is unable to make reliable inferences for tail and previously unseen items. Removing these ID features and their learned embeddings altogether to combat cold-start issue severely degrades the recommendation quality. Content-based item embeddings are more reliable, but they are expensive to store and use, particularly for users' past item interaction sequence. In this paper, we use Semantic IDs, a compact discrete item representations learned from content embeddings using RQ-VAE that captures hierarchy of concepts in items. We showcase how we use them as a replacement of item IDs in a resource-constrained ranking model used in an industrial-scale video sharing platform. Moreover, we show how Semantic IDs improves the generalization ability of our system, without sacrificing top-level metrics.

Siri is known to cut people off midsentence, but there's away to make Siri listen longer. CyberGuy shows you how to customize the wait time. I have a love /hate relationship with Siri. It shouldn't be so complicated and sure shouldn't sound like we need couples counseling. Unfortunately, this dysfunctional relationship I share with my iPhone's virtual assistant is systematic and widespread.

Linking latent features with their semantic meanings can improve the interpretability of recommender systems [16]. In recent years, many recommender systems have Typically, topic modeling techniques are utilized to extract utilized textual data for topic extraction to enhance topics from textual data and align extracted topics to latent interpretability. However, our findings reveal a noticeable features. For example, [13] introduces an approach that combines deficiency in the coherence of keywords the traditional latent factor model with Latent Dirichlet within topics, resulting in low explainability of the Allocation (LDA) [3] to uncover topics correlated with the model. This paper introduces a novel approach latent factors of both products and users. On the other hand, called entropy regularization to address the issue, [12] adopts a convolutional neural network (CNN) to encode leading to more interpretable topics extracted from textual reviews into item embeddings. The convolutional kernels recommender systems, while ensuring that the performance are then utilized to extract topics that correspond to the of the primary task stays competitively latent factors of items.

We study cross-modal recommendation of music tracks to be used as soundtracks for videos. This problem is known as the music supervision task. We build on a self-supervised system that learns a content association between music and video. In addition to the adequacy of content, adequacy of structure is crucial in music supervision to obtain relevant recommendations. We propose a novel approach to significantly improve the system's performance using structure-aware recommendation. The core idea is to consider not only the full audio-video clips, but rather shorter segments for training and inference. We find that using semantic segments and ranking the tracks according to sequence alignment costs significantly improves the results. We investigate the impact of different ranking metrics and segmentation methods.

If you've got an Amazon Echo speaker, you already know how much fun having Alexa at your fingertips can be despite the risk to your privacy and security. Whether playing music, answering random questions, or venting your frustrations, Alexa has become a part of your daily routine. Did you know, however, that it can do so much more? There are some seriously cool Alexa skills out there that can elevate your Echo speaker to a whole new level of intelligence. CLICK TO GET KURT'S FREE CYBERGUY NEWSLETTER WITH SECURITY ALERTS, QUICK TIPS, TECH REVIEWS AND EASY HOW-TO'S TO MAKE YOU SMARTER First, we have the "Ask My Buddy" skill, perfect for those who live alone or need assistance. With this skill, you can easily ask Alexa to notify a designated contact during an emergency, such as a fall or medical issue.