2026-10-26 ~ 11-01 · Personalization
Generative · Personalization · AI Search · Infra · Embodied AI
- 1. Exploratory creative pre-ranking: Hierarchical empirical-Bayes Thompson sampling for real-time creative optimization · 📖 Deep dive
AI-generated content makes it easy for advertisers to produce many creatives, but each gets very few impressions. This Roblox paper adds a creative pre-ranking layer between recall and ad ranking…
Appeared in: 10-29
- 2. Hierarchical GTV Estimation: Bridging the Gap between Container-Level Ranking and Item-Level Conversion · 📖 Deep dive
Local-services and live-commerce ads are ranked and billed per store or livestream room, but users actually buy one voucher or one product inside it, so directly predicting how much a click will sell gives a noisy…
Appeared in: 10-27
- 3. iFusion: Integrating Dynamic Interest Streams via Diffusion Model for Click-Through Rate Prediction · 📖 Deep dive
CTR models usually encode long-term and short-term behavior separately, then concatenate or gate them into one interest vector, so one-off noise in recent behavior, such as stocking up during a sale or buying for…
Appeared in: 10-27
- 4. Unified Heterogeneous Feature Interaction via Effective-Rank Scaling for Recommender Systems · 📖 Deep dive
Most production ranking models squeeze the user's behavior sequence into one fixed vector before it reaches the feature-interaction module, so later interaction layers can no longer see which clicks or add-to-carts…
Appeared in: 10-28
- 5. Rethinking Positional Encoding and Interaction Granularity for CTR Prediction · 📖 Deep dive
In unified Transformer CTR models like OneTrans, the authors tried six positional encodings on Kuaishou's KuaiRand data, and none beat using no positional encoding at all. Building on this…
Appeared in: 10-29
- 6. Towards a Unified Framework of Optimal Input-Side Dimensional Configuration for Deep CTR Prediction · 📖 Deep dive
There are three common ways to trim the input side of a CTR model: drop low-quality training samples, switch off useless features, and prune useless embedding dimensions. Earlier methods do one at a time…
Appeared in: 10-28
- 7. How Much Fresh Data Should CTR Recalibration Use? A Temporal Evaluation of Post-hoc Calibration · 📖 Deep dive
In ad systems, recalibrating a CTR model on the last few hours of fresh data is usually treated as a routine step that can only help. Testing it day by day on ten days of Avazu ad logs…
Appeared in: 10-28
- 8. MARS: Multi-resolution Adaptive Routing for Sequential Recommendation · 📖 Deep dive
To save compute, long-history recommenders compress a user's thousands of actions into a small set of cached vectors. This Duke and Meta paper finds that such caches barely remember a user's latest actions while very…
Appeared in: 10-27
- 9. BehaveRec: When Does Long-Term Behavioral History Help Sequential Recommendation? A Study with Session, History, and Community Signals
Are long-term history and community signals always useful? BehaveRec's answer is that it depends on how much history a user has. It combines three signals: a causal Transformer over the current session…
Appeared in: 10-29
- 10. Can Complementary Signals Bridge Similarity Islands? Manifold-Augmented Graph Embedding for Multimodal Recommendation
Multimodal recommenders often build item graphs from content similarity, so similar substitutes connect tightly while complements that are bought together but look different (a camera and a memory card) stay isolated…
Appeared in: 10-29