- 1. AGR: Agentic Generative Retrieval with Tool-Augmented Search · 📖 Deep dive
Generative retrieval need not encode every available feature up front. AGR requests intent, location, or behavioral evidence while decoding item identifiers, then resumes generation with a compact observation…
Appeared in: 10-21
- 2. Evolving Semantic Indexes for Generative Retrieval · 📖 Deep dive
An LLM evolves executable semantic-index programs while a tiny retriever screens them. With the target training recipe fixed, NQ320K Recall@100 rises from 85.82% to 90.57%.
Appeared in: 10-22
- 3. UniRec: A Unified Expressive-Aligned Generative Recommendation Framework for E-commerce · 📖 Deep dive
Generative recommendation generates a product's semantic IDs layer by layer, and before the first layer it has seen no product-side information at all; this is its gap with discriminative ranking models…
Appeared in: 10-24
- 4. OneLatent: Latent Reasoning for Efficient Foundation Recommendation Models · 📖 Deep dive
When a large language model serves as a recommendation foundation model and first writes a natural-language reasoning trace before outputting the next item's semantic ID, the reasoning text is far longer than the…
Appeared in: 10-24
- 5. Controllable Generative Recommendation via Guided Token Refinement · 📖 Deep dive
Generative recommenders emit item-ID tokens left to right, so temporary diversity at inference can only rewrite one position mid-decode and often produces IDs that do not exist in the catalog…
Appeared in: 10-23
- 6. GenPage: Towards End-to-End Generative Homepage Construction at Netflix · 📖 Deep dive
The Netflix homepage used to be assembled by a multi-stage recommender that retrieved candidate rows and titles, ranked them separately and then built the page; GenPage instead generates the whole page with one…
Appeared in: 10-24
- 7. Gryphon-v2: One Model in Place of a Cascade - Generate-and-Rank Recommender with Rollout Distillation · 📖 Deep dive
Gryphon-v2 encodes user history once, generates semantic item IDs, and ranks the resolved items with a small shared-state head. A larger teacher supplies training targets but stays out of serving…
Appeared in: 10-19
- 8. LightCRS: Lightweight Multi-Domain Conversational Recommendation System
LightCRS caches frozen multimodal product representations offline and uses a shared query-conditioned module to compare candidates online. HR@1 improves by 35.4% over the strongest baseline.
Appeared in: 10-22
- 9. From Habits to Discovery: Deploying LLMs for Personalized Generative Recommendations at Spotify
Spotify wants the home feed to surface podcasts the user has not yet formed a habit around (under 10 minutes on a show in the past 28 days). GLIDE, based on Llama 3.2 1B, first learns to translate between episode…
Appeared in: 10-23
- 10. Coarse-to-Fine Long-term Interest Modeling for Generative Recommendation · 📖 Deep dive
Generative recommendation generates the next item's semantic IDs one code at a time, and attention cost grows quadratically with sequence length, so these models usually take only the most recent few dozen behaviors…
Appeared in: 10-24