TOPIC

Recommender Architectures

39 synthesis notes · 42 source papers

View as

Do accuracy-optimized recommendations preserve user interest diversity?

Standard recommender systems rank by predicted relevance, which tends to saturate lists with the highest-confidence items. Does this approach naturally preserve the proportions of a user's multiple interests, or does it systematically crowd out smaller ones?

Why do accuracy-optimized recommenders crowd out minority interests?

Explores why recommendation models that maximize accuracy systematically over-represent a user's dominant interests while suppressing their lesser ones, even when both are measurable and real.

Can discrete codes transfer better than text embeddings?

Does inserting a discrete quantization layer between text and item representations improve cross-domain transfer in recommenders? This explores whether decoupling text from final embeddings reduces domain gap and text bias.

Can smaller models outperform their LLM teachers with enough data?

Explores whether student models trained on expanded teacher-generated labels can exceed teacher performance in production ranking tasks, and what data scale makes this possible.

Can model isolation solve streaming recommendation better than replay?

When continuously arriving user data arrives, does isolating parameters per task provide better control over forgetting old patterns while learning new ones than experience replay or knowledge distillation approaches?

Can simpler models beat deep networks for recommendation systems?

Does removing hidden layers and constraining self-similarity create a more effective collaborative filtering approach than deep autoencoders? This challenges the assumption that architectural depth drives performance.

Why do hash collisions hurt recommendation models so much?

Explores whether standard low-collision hashing works for embedding tables in recommenders, given that user and item frequencies follow power-law distributions rather than uniform ones.

Can single sessions alone rival history-rich recommendation?

Can encoder-only transformers with clever masking capture enough sequential signal from a single anonymous session to match recommenders that use extensive user history? This explores whether smart architecture can overcome sparse data.

Can a linear model beat deep collaborative filtering?

Does a shallow linear autoencoder with a zero-diagonal constraint outperform deeper neural models on collaborative filtering tasks? This challenges the field's assumption that depth and nonlinearity drive performance.

When can greedy bandits skip exploration entirely?

Under what conditions does natural randomness in incoming contexts eliminate the need for active exploration in contextual bandits? This matters for high-stakes domains like medicine where exploration carries real costs.

How can user vectors capture diverse interests without exploding in size?

Fixed-length user vectors compress all interests into one representation, losing information about varied tastes. Can we represent diverse interests efficiently without expanding dimensionality?

Can autoencoders solve the cold-start problem in recommendations?

Explores whether deep autoencoders combining collaborative filtering with side information can overcome the cold-start problem where new users or items lack rating history.

Can implicit feedback reveal both preference and confidence?

When users take implicit actions like purchases or watches, do those signals carry two separable pieces of information: what they prefer and how certain we should be? Explicit ratings can't make that distinction.

Can graphs unify collaborative filtering and side information?

How might merging user-item interactions with item attributes into a single graph structure allow recommendation systems to capture collaborative and attribute-based signals together, rather than separately?

Do LLM explanations faithfully describe their recommendation process?

When LLMs recommend items to groups, do their explanations match how they actually made the choice? This matters because users trust explanations to understand AI decision-making.

Can we distill LLM knowledge into graphs for real-time recommendations?

E-commerce needs sub-millisecond recommendations, but LLMs are too slow. Can we extract LLM insights offline into a knowledge graph that serves requests in production without sacrificing quality or explainability?

Can MLPs learn to match dot product similarity in practice?

Universal approximation theory suggests MLPs should learn any similarity function, including dot product. But does this theoretical promise hold up when training on real, finite datasets with practical constraints?

Why does dot product beat MLP-based similarity in practice?

Neural Collaborative Filtering theory suggests MLPs should outperform dot products as universal approximators. But what explains the empirical gap, and what role do data scale and deployment constraints play?

Why do ranking systems need to model selection bias explicitly?

Explores how training data from current rankers creates feedback loops that reinforce past decisions. Understanding this mechanism helps explain why naive approaches fail in production ranking systems.

Why does multinomial likelihood work better for click prediction?

Explores whether the choice of likelihood function—multinomial versus Gaussian or logistic—affects recommendation performance, and what structural properties make one better suited to modeling user clicks.

Why does multinomial likelihood work better for ranking recommendations?

Explores whether the choice of likelihood function in VAE-based collaborative filtering matters for matching training objectives to ranking evaluation metrics. Why items should compete for probability mass.

Why does Netflix use multiple ranking systems instead of one?

Netflix's homepage combines five distinct rankers optimizing different signals and time horizons. The question explores whether a single unified ranker could serve all user intents or if architectural separation is necessary.

What does Netflix need to optimize in those first 90 seconds?

Streaming users abandon after 60-90 seconds reviewing 1-2 screens. Does the recommender problem lie in predicting ratings accurately, or in making those limited screens immediately compelling?

How can real-time recommendations stay responsive and reproducible?

In-session signals improve ranking accuracy, but requiring fresh data during sessions forces real-time computation. This creates latency, network sensitivity, and debugging challenges that offset the relevance gains.

Do hash collisions really harm popular recommendation items?

Hash-based embedding tables assume uniform ID distribution, but real recommender systems show heavy-tailed frequency patterns. The question explores whether collisions actually concentrate damage on the high-traffic entities that matter most.

Why does collaborative filtering struggle with sparse user data?

Collaborative filtering datasets appear massive but hide a fundamental challenge: each user has rated only a tiny fraction of items. How does this per-user sparsity shape the modeling problem, and what techniques can overcome it?

How do feed ranking weights shape what content gets produced?

Feed-ranking weights are typically treated as neutral tuning parameters, but do they actually function as political levers that reshape producer behavior and the content supply itself?

Can reinforcement learning align summarization with ranking goals?

Generic LLM summaries optimize for readability, not ranking performance. Can training summarizers with downstream relevance scores as rewards fix this misalignment and produce summaries that actually help rankers match queries?

Can neural networks explore efficiently at recommendation scale?

Exploration—discovering unknown user preferences—normally requires expensive posterior uncertainty estimates. Can a neural architecture make Thompson sampling practical for real-world recommenders without prohibitive computational cost?

Why do recommendation systems miss recurring user preference patterns?

Most streaming recommendation systems treat preference changes as one-time drift events and discard old patterns. But user behavior often cycles—coffee shops on weekday mornings, gyms on weekends. How should systems account for these recurring periodicities instead of detecting and resetting against them?

Can graph structure patterns outperform direct edge signals in noisy data?

When user-behavior data is messy and unreliable, does looking at structural patterns across multiple edges produce better product recommendations than counting simple co-occurrences? This matters because e-commerce platforms need robust substitute graphs at billion-scale.

Why do global concept drift methods fail for recommender systems?

Recommender systems treat user preferences as individuals with distinct, asynchronous preference shifts. Can standard concept-drift approaches designed for population-level changes capture this per-user heterogeneity?

Can discretizing text embeddings improve recommendation transfer?

Does inserting a quantization step between text encodings and item representations reduce the recommender's over-reliance on text similarity and enable better cross-domain transfer?

Why do recommendation models fail when new users arrive?

Most recommendation algorithms are built assuming all users and items exist at training time. But real platforms constantly see new users and items. Can models be redesigned to handle unseen entities as a structural requirement?

Why do academic recommenders fail when deployed in production?

Academic recommendation models assume static test sets known at training time, but real platforms continuously receive new users, items, and interactions. Understanding this gap reveals what production systems actually need.

Can modeling multiple user personas improve recommendation accuracy?

Single-vector user representations compress all tastes into one place, potentially crowding out minority interests. Can representing users as multiple weighted personas adapt better to what's being scored and produce more accurate predictions?

Can attention mechanisms reveal which user taste explains each recommendation?

Single-vector user models collapse diverse tastes into one representation, losing expressiveness. Can weighting multiple personas by item relevance surface the right taste at the right time while making recommendations traceable?

Can one model handle both memorization and generalization?

Recommenders face a tradeoff between memorizing seen patterns and generalizing to new ones. Can a single architecture satisfy both needs without the cost of ensemble methods?

Can one model memorize and generalize better than two?

Does training memorization and generalization components jointly in a single model outperform training them separately and combining their predictions? This matters for building efficient recommendation systems that handle both rare and common user behaviors.

Source papers 42

The Arxiv papers behind this sub-topic. Links may take you off-site to arxiv.org.

Augmenting Netflix Search with In-Session Adapted Recommendations
“Machine Learning Ranker: We focus on building a machine learning model that can jointly optimize for the member’s long-term and short-term viewing preferences. To do so we took the following approach…
Calibrated Recommendations
When a user has watched, say, 70 romance movies and 30 action movies, then it is reasonable to expect the personalized list of recommended movies to be comprised of about 70% romance and 30% action mo…
Choosing the Right Weights: Balancing Value, Strategy, and Noise in Recommender Systems
Many recommender systems are based on optimizing a linear weighting of different user behaviors, such as clicks, likes, shares, etc. Though the choice of weights can have a significant impact, there i…
Collaborative Deep Learning for Recommender Systems
“To address the challenges above, we develop a hierarchical Bayesian model called collaborative deep learning (CDL) as a novel tightly coupled method for RS. We first present a Bayesian formulation of…
Collaborative Filtering Bandits
“Recommender Systems are an essential part of many successful on-line businesses, from e-commerce to on-line streaming, and beyond. Moreover, Computational Advertising can be seen as a recommendation …
Collaborative Filtering for Implicit Feedback Datasets
“An alternative strategy, our focus in this work, relies only on past user behavior without requiring the creation of explicit profiles. This approach is known as Collaborative Filtering (CF), a term …
Collaborative Filtering with Temporal Dynamics
“Modeling time drifting data is a central problem in data mining. Often, data is changing over time, and up to date modeling should be continuously updated to reflect its present nature. The analysis …
Consistent Explainers or Unreliable Narrators? Understanding LLM-generated Group Recommendations
Large Language Models (LLMs) are increasingly being implemented as joint decision-makers and explanation generators for Group Recommender Systems (GRS). In this paper, we evaluate these recommendation…
Content-aware Collaborative Music Recommendation Using Pre-trained Neural Networks
“Previous attempts on content-based music recommendation have achieved promising results. van den Oord et al. [13] utilize a neural network to map acoustic features to the song latent factors learned …
Deep Interest Network for Click-Through Rate Prediction
“Click-through rate prediction is an essential task in industrial applications, such as online advertising. Recently deep learning based models have been proposed, which follow a similar Embedding& ML…
Deep Neural Networks for YouTube Recommendations
“Recommending YouTube videos is extremely challenging from three major perspectives: Scale: Many existing recommendation algorithms proven to work well on small problems fail to operate on our scal…
Dynamically Expandable Graph Convolution for Streaming Recommendation
“Due to the real-world dynamics like user preference continuous shift and ever-increasing users and items, conventional recommender systems trained on the static fixed datasets usually suffer from: pr…
Embarrassingly Shallow Autoencoders for Sparse Data*
“Many recent improvements in collaborative filtering can be a attributed to deep learning approaches, e.g, [5, 7–9, 13, 21, 25, 26]. Unlike in areas like computer vision, however, it was found that a …
Enabling Explainable Recommendation in E-commerce with LLM-powered Product Knowledge Graph
How to leverage large language model’s superior capability in e-commerce recommendation has been a hot topic. In this paper, we propose LLM-PKG, an efficient approach that distills the knowledge of LL…
Explainable Recommendations via Attentive Multi-Persona Collaborative Filtering
“Collaborative Filtering (CF) methods model users’ taste and inclinations based on user-item interactions. Often, a user is represented by a single latent vector that encodes all of the user’s differe…
Factorization Meets the Neighborhood: a Multifaceted Collaborative Filtering Model
Recommender systems provide users with personalized suggestions for products or services. These systems often rely on Collaborating Filtering (CF), where past transactions are analyzed in order to est…
GHRS: Graph-based Hybrid Recommendation System with Application to Movie Recommendation
“RS methods are mainly categorized into Collaborative Filtering (CF), Content-Based Filtering (CBF), and hybrid recommender system based on the input data (Adomavicius and Tuzhilin, 2005). CF models (…
HyperBandit: Contextual Bandit with Hypernetwork for Time-Varying User Preferences in Streaming Recommendation
“While the demand for personalized recommendations has increased due to the growth of online platforms and user-generated content, it is crucial to emphasize that the recommendation models need to be …
InTune: Reinforcement Learning-based Data Pipeline Optimization for Deep Recommendation Models
“Recommendation systems now underpin many essential components of the web ecosystem, including search result ranking, ecommerce product placement, and media suggestions in streaming services. Over the…
KGAT: Knowledge Graph Attention Network for Recommendation
“The success of recommendation system makes it prevalent in Web applications, ranging from search engines, E-commerce, to social media sites and news portals — without exaggeration, almost every servi…
Large Scale Product Graph Construction for Recommendation in E-commerce
“On the other hand, Taobao has an enormous amount of real user behavior data from billions of customers, which are much stronger and reliable signals for capturing product relationships than text data…
Learning Distributed Representations from Reviews for Collaborative Filtering
“Collaborative filtering has been successfully used for recommendation systems (see, e.g., [17]). A typical approach to using collaborative filtering for recommendation systems is to consider all the …
Learning Vector-Quantized Item Representation for Transferable Sequential Recommenders
“Sequential recommender systems have been widely deployed on various application platforms for recommending items of interest to users. Typically, such a recommendation task is formulated as a sequenc…
Learning to Rank for Recommender Systems
Recommender system aim at providing a personalized list of items ranked according to the preferences of the user, as such ranking methods are at the core of many recommendation algorithms. The topic o…
Lessons Learnt From Consolidating ML Models in a Large Scale Recommendation System
![A screenshot of a diagram](/assets/paper-images/LargeActionModelsFromInceptionToImplementation.png)
Methodologies for Improving Modern Industrial Recommender Systems
Recommender system (RS) is an established technology with successful applications in social media, e-commerce, entertainment, and more. RSs are indeed key to the success of many popular APPs, such as …
Monolith: Real Time Recommendation System With Collisionless Embedding Table
“The past decade witnessed a boom of businesses powered by recommendation techniques. In pursuit of a better customer experience, delivering personalized content for each individual user as real-time …
Mostly Exploration-Free Algorithms for Contextual Bandits
“However, exploration may be prohibitively costly or infeasible in a variety of practical environments (Bird et al. 2016). In medical decision-making, choosing a treatment that is not the estimated-be…
Neural Collaborative Filtering
“In the era of information explosion, recommender systems play a pivotal role in alleviating information overload, having been widely adopted by many online services, including E-commerce, online news…
Neural Collaborative Filtering vs. Matrix Factorization Revisited
“Embedding based models have been the state of the art in collaborative filtering for over a decade. Traditionally, the dot product or higher order equivalents have been used to combine two or more em…
Optimizing Encoder-Only Transformers for Session-Based Recommendation Systems
Session-based recommendation is the task of predicting the next item a user will interact with, often without access to historical user data. In this work, we introduce Sequential Masked Modeling, a n…
Recommender Systems with Social Regularization
“Traditional recommender systems always ignore social relationships among users. But in our real life, when we are asking our friends for recommendations of nice digital cameras or touching movies, we…
Recommending What Video to Watch Next: A Multitask Ranking System
“In this paper, we introduce a large scale multi-objective ranking system for recommending what video to watch next on an industrial video sharing platform. The system faces many real-world challenges…
Scalable Neural Contextual Bandit for Recommender Systems
High-quality recommender systems ought to deliver both innovative and relevant content through effective and exploratory interactions with users. Yet, supervised learning-based neural networks, which …
Situating Recommender Systems in Practice: Towards Inductive Learning and Incremental Updates
“With information systems becoming larger scale, recommendation systems are a topic of growing interest in machine learning research and industry. Even though progress on improving model design has be…
The Netflix Recommender System: Algorithms, Business Value, and Innovation
“Internet TV is about choice: what to watch, when to watch, and where to watch, compared with linear broadcast and cable systems that offer whatever is now playing on perhaps 10 to 20 favorite channel…
Tube2Vec: Social and Semantic Embeddings of YouTube Channels
Research using YouTube data often explores social and semantic dimensions of channels and videos. Typically, analyses rely on laborious manual annotation of content and content creators, often found b…
Unifying Nearest Neighbors Collaborative Filtering
We study collaborative filtering for applications in which there exists for every user a set of items about which the user has given binary, positive-only feedback (one-class collaborative filtering).…
Using Navigation to Improve Recommendations in Real-Time
Implicit feedback is a key source of information for many recommendation and personalization approaches. However, using it typically requires multiple episodes of interaction and roundtrips to a recom…
Variational Autoencoders for Collaborative Filtering
“Collaborative filtering is among the most widely applied approaches in recommender systems. Collaborative filtering predicts what items a user will prefer by discovering and exploiting the similarity…
Wide & Deep Learning for Recommender Systems
“A recommender system can be viewed as a search ranking system, where the input query is a set of user and contextual information, and the output is a ranked list of items. Given a query, the recommen…
“What do others think?”: Task-Oriented Conversational Modeling with Subjective Knowledge
Task-oriented Dialogue (TOD) Systems aim to build dialogue systems that assist users in accomplishing specific goals, such as booking a hotel or a restaurant. Traditional TODs rely on domain-specific …