The Hallway Track
9 tracked signals on recommendation-systems.
Spotify rebuilt its recommendation system as LLM-centric 'generative personalization' at 760M user scale
“moving from personalization as guesswork...to personalization as reasoning, which is able to analyze those results and really understand whether they are relevant to that user in a given context”
Meta's GenRecs Tokenization applies LLM semantic vocabulary to ads, yielding +0.35% conversion lift
“A newly launched "trail shoe" inherits what the model has already learned about that category, rather than starting a new representation with no signal.”
Meta's unified GEM foundation model delivers 18.9% conversion and 12% CTR improvement across all ad surfaces
“18.9% conversion improvement and 12% ad click-through rate (CTR) improvement”
Courts ruled Meta's addictive AI-driven app design—not just content—causes teen harm.
“Just because of the features alone that drive addiction, these companies can be held accountable. That's a huge message for these companies.”
Meta doubled GEM ads model training efficiency to 20-25% MFU while scaling FLOPs 4x in 12 months
Meta achieves LLM-style scaling laws for ads ranking with 6% Instagram conversion lift
“a flexible production strategy that helps generalize sequence learning models and establish an LLM-style scaling law that predictably balances model performance with compute”
Meta's SilverTorch unifies recommendation retrieval into one neural network with 23.7x throughput gains
NVIDIA Dynamo-Triton enables deployment of HSTU generative recommender systems at scale
AWS presents a multi-tower neural network NBP recommendation system with per-customer explainability for banking.