This special edition is part of our Guest Prudie series, where we ask smart, thoughtful people to step in as Prudie for the day and give you advice. Today's columnist is number one New York Times-bestselling author Leigh Bardugo. She is the author of the books The Familiar, Ninth House and the creator of the Grishaverse (now a Netflix original series) which spans the Shadow and Bone trilogy, the Six of Crows duology, the King of Scars duology. Her short fiction has appeared in multiple anthologies including The Best American Science Fiction and Fantasy. She lives in Los Angeles and is an associate fellow of Pauli Murray College at Yale University. We asked Bardugo to weigh in on "romantic" gestures gone wrong, conversational vampires, and vocal dogs: I recently met a man on a dating app. We hit it off quickly. We were texting all of the time about work, writing, and the world--often getting pretty flirty. I was having tons of fun. He was charming and seemed to me conspicuously brilliant.

This paper presents ReasoningRec, a reasoning-based recommendation framework that leverages Large Language Models (LLMs) to bridge the gap between recommendations and human-interpretable explanations. In contrast to conventional recommendation systems that rely on implicit user-item interactions, ReasoningRec employs LLMs to model users and items, focusing on preferences, aversions, and explanatory reasoning. The framework utilizes a larger LLM to generate synthetic explanations for user preferences, subsequently used to fine-tune a smaller LLM for enhanced recommendation accuracy and human-interpretable explanation. Our experimental study investigates the impact of reasoning and contextual information on personalized recommendations, revealing that the quality of contextual and personalized data significantly influences the LLM's capacity to generate plausible explanations. Empirical evaluations demonstrate that ReasoningRec surpasses state-of-the-art methods by up to 12.5\% in recommendation prediction while concurrently providing human-intelligible explanations. The code is available here: https://github.com/millenniumbismay/reasoningrec.

Recent advancements in recommender systems have focused on leveraging Large Language Models (LLMs) to improve user preference modeling, yielding promising outcomes. However, current LLM-based approaches struggle to fully leverage user behavior sequences, resulting in suboptimal preference modeling for personalized recommendations. In this study, we propose a novel Counterfactual Fine-Tuning (CFT) method to address this issue by explicitly emphasizing the role of behavior sequences when generating recommendations. Specifically, we employ counterfactual reasoning to identify the causal effects of behavior sequences on model output and introduce a task that directly fits the ground-truth labels based on these effects, achieving the goal of explicit emphasis. Additionally, we develop a token-level weighting mechanism to adjust the emphasis strength for different item tokens, reflecting the diminishing influence of behavior sequences from earlier to later tokens during predicting an item. Extensive experiments on real-world datasets demonstrate that CFT effectively improves behavior sequence modeling. Our codes are available at https://github.com/itsmeyjt/CFT.

Sequential recommender systems (SRSs) aim to predict the subsequent items which may interest users via comprehensively modeling users' complex preference embedded in the sequence of user-item interactions. However, most of existing SRSs often model users' single low-level preference based on item ID information while ignoring the high-level preference revealed by item attribute information, such as item category. Furthermore, they often utilize limited sequence context information to predict the next item while overlooking richer inter-item semantic relations. To this end, in this paper, we proposed a novel hierarchical preference modeling framework to substantially model the complex low- and high-level preference dynamics for accurate sequential recommendation. Specifically, in the framework, a novel dual-transformer module and a novel dual contrastive learning scheme have been designed to discriminatively learn users' low- and high-level preference and to effectively enhance both low- and high-level preference learning respectively. In addition, a novel semantics-enhanced context embedding module has been devised to generate more informative context embedding for further improving the recommendation performance. Extensive experiments on six real-world datasets have demonstrated both the superiority of our proposed method over the state-of-the-art ones and the rationality of our design.

What we once called the Google Chromecast (and then the Chromecast with Google TV) is now the Google TV Streamer. I won't pretend to understand the reasoning behind any product's rebrand, but at least this one makes a bit of sense. Casting content from elsewhere used to be a big reason TV dongles existed. Today, streaming devices primarily provide the brains required to watch content from Netflix, Disney and other streaming services on almost any screen, and casting is a bit of an afterthought. A name that focuses on Google TV's interface instead of casting seems right in 2024.

Personalized recommendation is a ubiquitous application on the internet, with many industries and hyperscalers extensively leveraging Deep Learning Recommendation Models (DLRMs) for their personalization needs (like ad serving or movie suggestions). With growing model and dataset sizes pushing computation and memory requirements, GPUs are being increasingly preferred for executing DLRM inference. However, serving newer DLRMs, while meeting acceptable latencies, continues to remain challenging, making traditional deployments increasingly more GPU-hungry, resulting in higher inference serving costs. In this paper, we show that the embedding stage continues to be the primary bottleneck in the GPU inference pipeline, leading up to a 3.2x embedding-only performance slowdown. To thoroughly grasp the problem, we conduct a detailed microarchitecture characterization and highlight the presence of low occupancy in the standard embedding kernels. By leveraging direct compiler optimizations, we achieve optimal occupancy, pushing the performance by up to 53%. Yet, long memory latency stalls continue to exist. To tackle this challenge, we propose specialized plug-and-play-based software prefetching and L2 pinning techniques, which help in hiding and decreasing the latencies. Further, we propose combining them, as they complement each other. Experimental evaluations using A100 GPUs with large models and datasets show that our proposed techniques improve performance by up to 103% for the embedding stage, and up to 77% for the overall DLRM inference pipeline.

Recent advancements in diffusion models have shown promising results in sequential recommendation (SR). However, current diffusion-based methods still exhibit two key limitations. First, they implicitly model the diffusion process for target item embeddings rather than the discrete target item itself, leading to inconsistency in the recommendation process. Second, existing methods rely on either implicit or explicit conditional diffusion models, limiting their ability to fully capture the context of user behavior and leading to less robust target item embeddings. In this paper, we propose the Dual Conditional Diffusion Models for Sequential Recommendation (DCRec), introducing a discrete-to-continuous sequential recommendation diffusion framework. Our framework introduces a complete Markov chain to model the transition from the reversed target item representation to the discrete item index, bridging the discrete and continuous item spaces for diffusion models and ensuring consistency with the diffusion framework. Building on this framework, we present the Dual Conditional Diffusion Transformer (DCDT) that incorporates the implicit conditional and the explicit conditional for diffusion-based SR. Extensive experiments on public benchmark datasets demonstrate that DCRec outperforms state-of-the-art methods.

Large-scale industrial recommendation models predict the most relevant items from catalogs containing millions or billions of options. To train these models efficiently, a small set of irrelevant items (negative samples) is selected from the vast catalog for each relevant item (positive example), helping the model distinguish between relevant and irrelevant items. Choosing the right negative sampling method is a common challenge. We address this by implementing and comparing various negative sampling methods - random, popularity-based, in-batch, mixed, adaptive, and adaptive with mixed variants - on modern sequential recommendation models. Our experiments, including hyperparameter optimization and 20x repeats on three benchmark datasets with varying popularity biases, show how the choice of method and dataset characteristics impact key model performance metrics. We also reveal that average performance metrics often hide imbalances across popularity bands (head, mid, tail). We find that commonly used random negative sampling reinforces popularity bias and performs best for head items. Popularity-based methods (in-batch and global popularity negative sampling) can offer balanced performance at the cost of lower overall model performance results. Our study serves as a practical guide to the trade-offs in selecting a negative sampling method for large-scale sequential recommendation models. Code, datasets, experimental results and hyperparameters are available at: https://github.com/apple/ml-negative-sampling.

After a long wait, Apple has finally released its artificial intelligence (AI) tools for iPhone - to a select few. Apple Intelligence, a suite of AI tools announced in June, became available to owners of some iPhones around the world on Monday. The new features include notification summaries, tools to assist users in writing messages, and a glowing new interface for virtual assistant Siri. But they will only be available to people with the latest devices - including all iPhone 16 models, and the iPhone 15 Pro and Pro Max. Apple Intelligence is also available on Mac computers and iPad tablets that are powered by its latest chips. But some of the tools made available on Monday have arrived later than equivalent features on other popular devices.

Apple is set to launch its new iOS 18.1 that will include the long-awaited AI feature and several security fixes for iPhones running on the previous version. CEO Tim Cook has touted Apple Intelligence as'a new chapter of innovation,' focusing on'generative' AI models that enable users to create text or images from prompts. The system will have the ability to create'Genmojis,' new emoji characters based on text prompts in iMessage, edit photos and include a revamped Siri with better conversational skills. The new iOS 18 system is set to hit smartphones at 1pm ET, but only the iPhone 16 family and high-end 15 devices support Apple Intelligence. There is also a waitlist for the AI feature and users can claim their spot after downloading the update.