The rapidly evolving landscape of products, surfaces, policies, and regulations poses significant challenges for deploying state-of-the-art recommendation models at industry scale, primarily due to data fragmentation across domains and escalating infrastructure costs that hinder sustained quality improvements.

Recent advances in recommendation scaling laws have led to foundation models of unprecedented complexity. While these models offer superior performance, their computational demands make real-time serving impractical, often forcing practitioners to rely on knowledge distillation—compromising serving quality for efficiency. To address this challenge, we present SOLARIS (Speculative Offloading of Latent-bAsed Representation for Inference Scaling), a novel framework inspired by speculative decoding. SOLARIS proactively precomputes user-item interaction embeddings by predicting which user-item pairs are likely to appear in future requests, and asynchronously generating their foundation model representations ahead of time. This approach decouples the costly foundation model inference from the latency-critical serving path, enabling real-time knowledge transfer from models previously considered too expensive for online use. Deployed across Meta's advertising system serving billions of daily requests, SOLARIS achieves 0.67% revenue-driving top-line metrics gain, demonstrating its effectiveness at scale.

Scaling laws play an instrumental role in the sustainable improvement in model quality. Unfortunately, recommendation models to date do not exhibit such laws similar to those observed in the domain of large language models, due to the inefficiencies of their upscaling mechanisms. This limitation poses significant challenges in adapting these models to increasingly more complex real-world datasets. In this paper, we propose an effective network architecture based purely on stacked factorization machines, and a synergistic upscaling strategy, collectively dubbed Wukong, to establish a scaling law in the domain of recommendation. Wukong’s unique design makes it possible to capture diverse, any-order of interactions simply through taller and wider layers. We conducted extensive evaluations on six public datasets, and our results demonstrate that Wukong consistently outperforms state-of-the-art models quality-wise. Further, we assessed Wukong’s scalability on an internal, large-scale dataset. The results show that Wukong retains its superiority in quality over state-of-the-art models, while holding the scaling law across two orders of magnitude in model complexity, extending beyond 100 GFLOP/example, where prior arts fall short.

In online advertising market, advertisers aim to target the right audience through media channels such as display, mobile, video, or social with limited budget. Usually the advertiser can collect high valuable users from their existing customs, whose size however is usually small. Hence, it is very important for advertisers to reach more audiences who can potentially be turned into future customers. How to induce the preferable attributes from the seed audience and hence identify more similar users from billions of cookie-able users within the runtime of a few minutes is a challenge to data science techniques. This is an important problem and many researchers have proposed several solutions [1]-[6] In this paper, we propose an efficient approach which can return a set of Boolean rules that defines the extended audience by analyzing the seed audience. As an additional feature, the proposed system also provides an option for advertisers to participate in the decision making process which could potentially improve the user reachability as well as the campaign performance. The experiment results show the superior performance of the proposed system compared with other methods in terms of efficiency, user reachability, action rate, click through rate, etc.