Guided Flows for Generative Modeling and Decision Making
This work addresses performance and efficiency bottlenecks in generative modeling and decision-making tasks, offering incremental improvements by adapting an existing technique to a new model type.
The paper integrates classifier-free guidance into Flow Matching models to improve conditional generative modeling, achieving state-of-the-art performance in image generation and text-to-speech synthesis, and demonstrates a 10x speedup in offline reinforcement learning plan generation compared to diffusion models.
Classifier-free guidance is a key component for enhancing the performance of conditional generative models across diverse tasks. While it has previously demonstrated remarkable improvements for the sample quality, it has only been exclusively employed for diffusion models. In this paper, we integrate classifier-free guidance into Flow Matching (FM) models, an alternative simulation-free approach that trains Continuous Normalizing Flows (CNFs) based on regressing vector fields. We explore the usage of \emph{Guided Flows} for a variety of downstream applications. We show that Guided Flows significantly improves the sample quality in conditional image generation and zero-shot text-to-speech synthesis, boasting state-of-the-art performance. Notably, we are the first to apply flow models for plan generation in the offline reinforcement learning setting, showcasing a 10x speedup in computation compared to diffusion models while maintaining comparable performance.