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

Google Research introduced Retrieve-for-Train (R4T), a search framework that trains a fan-out language model with reinforcement learning once, then uses that model to synthesize training data for a 53.9M-parameter diffusion retriever that generates all retrieval directions in a single pass, running 12× to 20× faster than autoregressive fan-out. The fan-out language model is trained with groundedness, diversity, and alignment rewards to return coherent, diverse result sets. No code or model weights have been released yet.