State of the Field
Recent advancements in recommendation systems are increasingly focused on enhancing model efficiency and accuracy while addressing the complexities of user interactions. A notable trend is the development of scalable architectures that optimize feature interactions without incurring excessive latency, as seen in recent work that employs multi-stage frameworks and efficient ranking models. These innovations are particularly relevant for platforms like TikTok and Facebook, where user engagement metrics are critical. Additionally, there is a growing emphasis on incorporating diverse data types, such as sensory attributes extracted from product reviews, to enrich item representations and improve recommendation relevance. The field is also tackling challenges related to shared accounts, with new methods designed to disentangle user behaviors and infer latent preferences more effectively. Overall, the push towards more sophisticated, user-centric models is poised to address commercial challenges in personalization and engagement, making recommendations more relevant and efficient across various applications.
Papers
1–10 of 11LLaTTE: Scaling Laws for Multi-Stage Sequence Modeling in Large-Scale Ads Recommendation
We present LLaTTE (LLM-Style Latent Transformers for Temporal Events), a scalable transformer architecture for production ads recommendation. Through systematic experiments, we demonstrate that sequen...
Zenith: Scaling up Ranking Models for Billion-scale Livestreaming Recommendation
Accurately capturing feature interactions is essential in recommender systems, and recent trends show that scaling up model capacity could be a key driver for next-level predictive performance. While ...
When Text-as-Vision Meets Semantic IDs in Generative Recommendation: An Empirical Study
Semantic ID learning is a key interface in Generative Recommendation (GR) models, mapping items to discrete identifiers grounded in side information, most commonly via a pretrained text encoder. Howev...
From Bias to Balance: Fairness-Aware Paper Recommendation for Equitable Peer Review
Despite frequent double-blind review, systemic biases related to author demographics still disadvantage underrepresented groups. We start from a simple hypothesis: if a post-review recommender is trai...
Multi-Behavior Sequential Modeling with Transition-Aware Graph Attention Network for E-Commerce Recommendation
User interactions on e-commerce platforms are inherently diverse, involving behaviors such as clicking, favoriting, adding to cart, and purchasing. The transitions between these behaviors offer valuab...
Sensory-Aware Sequential Recommendation via Review-Distilled Representations
We propose a novel framework for sensory-aware sequential recommendation that enriches item representations with linguistically extracted sensory attributes from product reviews. Our approach, \textsc...
DisenReason: Behavior Disentanglement and Latent Reasoning for Shared-Account Sequential Recommendation
Shared-account usage is common on streaming and e-commerce platforms, where multiple users share one account. Existing shared-account sequential recommendation (SSR) methods often assume a fixed numbe...
MALLOC: Benchmarking the Memory-aware Long Sequence Compression for Large Sequential Recommendation
The scaling law, which indicates that model performance improves with increasing dataset and model capacity, has fueled a growing trend in expanding recommendation models in both industry and academia...
Position-Aware Sequential Attention for Accurate Next Item Recommendations
Sequential self-attention models usually rely on additive positional embeddings, which inject positional information into item representations at the input. In the absence of positional signals, the a...
AlphaFree: Recommendation Free from Users, IDs, and GNNs
Can we design effective recommender systems free from users, IDs, and GNNs? Recommender systems are central to personalized content delivery across domains, with top-K item recommendation being a fund...