distillation
14 talks
Distill the LLM, Don't Serve It: Search & Personalization at DoorDash
Bringing Continual Learning into Enterprises
Data Quality Is the Compute Multiplier
You Might Not Need 50 Diffusion Steps
Stop Making Models Bigger, Make Them Behave
How We Built Zeta2: Training an Edit Prediction Model in Production
Building Generative Image & Video Models at Scale
Latent Space Paper Club: AIEWF Special Edition (Test of Time, DeepSeek R1/V3)
OpenThoughts: Data Recipes for Reasoning Models
360Brew: LLM-based Personalized Ranking and Recommendation
Improving Recommendation Systems and Search in the Age of LLMs
One Model to Rule Netflix Recommendations
What We Learned from Using LLMs in Pinterest
LLM Recommendation Systems (RecSys)