Amazon has begun what seems to be a very slow rollout for Alexa, its new AI-enhanced version of Alexa. But if you're looking for hands-on impressions from everyday folks who've actually tried the new Alexa, they're curiously hard to find. During Amazon's earnings call late last week, CEO Andy Jassy told investors that "over 100,000" users are currently testing Alexa, which Amazon unveiled more than two months ago during a splashy New York City press event. "People are really liking Alexa so far," Jassy reportedly said on the earnings call. A hundred thousand people may sound like a lot, but as AFTVnews points out, it's just a tiny fraction of Alexa's estimated 100 million-plus active user base, meaning the vast majority of Alexa users has yet to try Alexa out for themselves. Meanwhile, the lucky few who have kicked the tires on the new AI Alexa are generally keeping mum about it.

User sequence modeling is crucial for modern large-scale recommendation systems, as it enables the extraction of informative representations of users and items from their historical interactions. These user representations are widely used for a variety of downstream tasks to enhance users' online experience. A key challenge for learning these representations is the lack of labeled training data. While self-supervised learning (SSL) methods have emerged as a promising solution for learning representations from unlabeled data, many existing approaches rely on extensive negative sampling, which can be computationally expensive and may not always be feasible in real-world scenario. In this work, we propose an adaptation of Barlow Twins, a state-of-the-art SSL methods, to user sequence modeling by incorporating suitable augmentation methods. Our approach aims to mitigate the need for large negative sample batches, enabling effective representation learning with smaller batch sizes and limited labeled data. We evaluate our method on the MovieLens-1M, MovieLens-20M, and Yelp datasets, demonstrating that our method consistently outperforms the widely-used dual encoder model across three downstream tasks, achieving an 8%-20% improvement in accuracy. Our findings underscore the effectiveness of our approach in extracting valuable sequence-level information for user modeling, particularly in scenarios where labeled data is scarce and negative examples are limited.

Large Language Model (LLM)-based recommendation systems leverage powerful language models to generate personalized suggestions by processing user interactions and preferences. Unlike traditional recommendation systems that rely on structured data and collaborative filtering, LLM-based models process textual and contextual information, often using cloud-based infrastructure. This raises privacy concerns, as user data is transmitted to remote servers, increasing the risk of exposure and reducing control over personal information. To address this, we propose a hybrid privacy-preserving recommendation framework which separates sensitive from nonsensitive data and only shares the latter with the cloud to harness LLM-powered recommendations. To restore lost recommendations related to obfuscated sensitive data, we design a de-obfuscation module that reconstructs sensitive recommendations locally. Experiments on real-world e-commerce datasets show that our framework achieves almost the same recommendation utility with a system which shares all data with an LLM, while preserving privacy to a large extend. Compared to obfuscation-only techniques, our approach improves HR@10 scores and category distribution alignment, offering a better balance between privacy and recommendation quality. Furthermore, our method runs efficiently on consumer-grade hardware, making privacy-aware LLM-based recommendation systems practical for real-world use.

Accurately modeling user preferences is vital not only for improving recommendation performance but also for enhancing transparency in recommender systems. Conventional user profiling methods, such as averaging item embeddings, often overlook the evolving, nuanced nature of user interests, particularly the interplay between short-term and long-term preferences. In this work, we leverage large language models (LLMs) to generate natural language summaries of users' interaction histories, distinguishing recent behaviors from more persistent tendencies. Our framework not only models temporal user preferences but also produces natural language profiles that can be used to explain recommendations in an interpretable manner. These textual profiles are encoded via a pre-trained model, and an attention mechanism dynamically fuses the short-term and long-term embeddings into a comprehensive user representation. Beyond boosting recommendation accuracy over multiple baselines, our approach naturally supports explainability: the interpretable text summaries and attention weights can be exposed to end users, offering insights into why specific items are suggested. Experiments on real-world datasets underscore both the performance gains and the promise of generating clearer, more transparent justifications for content-based recommendations.

Graph convolutional networks have recently gained prominence in collaborative filtering (CF) for recommendations. However, we identify potential bottlenecks in two foundational components. First, the embedding layer leads to a latent space with limited capacity, overlooking locally observed but potentially valuable preference patterns. Also, the widely-used neighborhood aggregation is limited in its ability to leverage diverse preference patterns in a fine-grained manner. Building on spectral graph theory, we reveal that these limitations stem from graph filtering with a cut-off in the frequency spectrum and a restricted linear form. To address these issues, we introduce ChebyCF, a CF framework based on graph spectral filtering. Instead of a learned embedding, it takes a user's raw interaction history to utilize the full spectrum of signals contained in it. Also, it adopts Chebyshev interpolation to effectively approximate a flexible non-linear graph filter, and further enhances it by using an additional ideal pass filter and degree-based normalization. Through extensive experiments, we verify that ChebyCF overcomes the aforementioned bottlenecks and achieves state-of-the-art performance across multiple benchmarks and reasonably fast inference. Our code is available at https://github.com/chanwoo0806/ChebyCF.

Augmenting specialised machine learning techniques into traditional graph learning models has achieved notable success across various domains, including federated graph learning, dynamic graph learning, and graph transformers. However, the intricate mechanisms of these specialised techniques introduce significant challenges in maintaining model fairness, potentially resulting in discriminatory outcomes in high-stakes applications such as recommendation systems, disaster response, criminal justice, and loan approval. This paper systematically examines the unique fairness challenges posed by Graph Learning augmented with Machine Learning (GL-ML). It highlights the complex interplay between graph learning mechanisms and machine learning techniques, empha-sising how the augmentation of machine learning both enhances and complicates fairness. Additionally, we explore four critical techniques frequently employed to improve fairness in GL-ML methods. By thoroughly investigating the root causes and broader implications of fairness challenges in this rapidly evolving field, this work establishes a robust foundation for future research and innovation in GL-ML fairness.

One day after Meta rolled out its standalone AI app, Mark Zuckerberg has shared more about how the company plans to eventually monetize its generative AI assistant. During the company's first quarter earnings call, Zuckerberg said Meta AI could one day show ads and product recommendations. He also hinted at plans for a subscription component for those who want a more "premium" version of the assistant. "I think that there will be a large opportunity to show product recommendations or ads, as well as a premium service for people who want to unlock more compute for additional functionality or intelligence," Zuckerberg said. He added that for now the company is more focused on growing Meta AI's usage.

In September 2023, Adam Raines made a Reddit post revealing what feels like a near-universal problem for singles: His dating app conversations are painfully boring. Attached to the post, titled "Sometimes, texting on dating apps feelings (sic) like hitting your head against a brick wall," is a screenshot of a bone-dry Tinder conversation between him and one of his matches, in which Raines' curiosity is met with short, dead-ended answers. "The vast majority of my online dating interactions have been like that," says Raines, 25, a gay man living in the UK who asked to use a pseudonym to protect his privacy. Many users in the thread echoed his sentiment and offered explanations or theories as to why conversations on dating apps are often unsatisfying. "I see I'm not the only one getting that type of energy lol," one wrote, as another noted, "It sucks, and if people swiped more mindfully this wouldn't happen, but a lot of guys are so beaten down by the dating app experience they feel like they don't have any other choice and want whatever validation they can get."

Tapple, a popular Japanese dating app, on Wednesday launched a new feature that allows users to verify their unmarried status via their government-issued identification My Number card -- a first in the country. With the app boasting over 20 million users, worries about dishonest relationship statuses were common. According to a survey conducted by Tapple, over half of the nearly 5,500 respondents -- 53.8% of men and 68.6% of women -- have previously expressed concerns about whether the people they meet on the app might be married. The survey also found that nearly all female respondents and over 80% of male respondents said they want proof the people they match with are single. Similarly, around 80% of respondents expressed an interest in being able to verify their own single status to ease any doubts others may have about their availability.

As new products are emerging daily, recommendation systems are required to quickly adapt to possible new domains without needing extensive retraining. This work presents ``X-Cross'' -- a novel cross-domain sequential-recommendation model that recommends products in new domains by integrating several domain-specific language models; each model is fine-tuned with low-rank adapters (LoRA). Given a recommendation prompt, operating layer by layer, X-Cross dynamically refines the representation of each source language model by integrating knowledge from all other models. These refined representations are propagated from one layer to the next, leveraging the activations from each domain adapter to ensure domain-specific nuances are preserved while enabling adaptability across domains. Using Amazon datasets for sequential recommendation, X-Cross achieves performance comparable to a model that is fine-tuned with LoRA, while using only 25% of the additional parameters. In cross-domain tasks, such as adapting from Toys domain to Tools, Electronics or Sports, X-Cross demonstrates robust performance, while requiring about 50%-75% less fine-tuning data than LoRA to make fine-tuning effective. Furthermore, X-Cross achieves significant improvement in accuracy over alternative cross-domain baselines. Overall, X-Cross enables scalable and adaptive cross-domain recommendations, reducing computational overhead and providing an efficient solution for data-constrained environments.