Google Research Introduces Retrieve-for-Train (R4T): An RL-Compiled Diffusion Retriever for 12× to 20× Faster Query Fan-Out

MarkTechPost · 4 weeks ago · 35 ● 2 sources

Google Research introduced Retrieve-for-Train (R4T) to generate diverse retrieval result sets from one query instead of near-duplicate matches. The R4T diffusion retriever runs in 0.07 seconds at batch size 8, versus about 1.46 seconds for an autoregressive fan-out. R4T first learns retrieval fan-out with offline reinforcement learning and then deploys a 53.9M-parameter diffusion model for a single-pass, faster retrieval fan-out.