• context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.

Netflix collapsed a multi-stage row-and-entity ranking pipeline into one autoregressive model that emits the page piece by piece, each choice conditioned on prior ones — an applied case of prompt-response modeling replacing hand-built recommendation stages.

• blogIn-House LLM Serving at Netflix
• blogGenRec: Towards LLM-Native Recommendation at Netflix

• blogGenRec: Towards LLM-Native Recommendation at NetflixNetflix Technology Blog
• articleGenerative RecommendationsShopify Engineering
• articleHow Generative Recommenders Are Redefining RecSys at ScaleElizabeth Goodman

Checking sign-in…

Loading comments…