How to Guide Your Language Flow

Summary

Probe guidance, a novel method for guiding flow matching models, is introduced. It uses the frozen internal states of an existing diffusion model to construct a guidance signal, operating on a similar principle as autoguidance but removing the need for an additional forward pass at inference time. The method provides a reliable path ensuring the weak and strong models share similar dynamics. Probe guidance sets a new state-of-the-art performance on unconditional generation when benchmarked on continuous diffusion language models. Applied to a 1.7B diffusion language model, it consistently improves multiple choice question answering benchmarks.

Takeaways

1. 1

   Probe guidance uses frozen internal states of an existing diffusion model to construct a guidance signal, eliminating the need for an additional forward pass at inference time.

2. 2

   When applied to a 1.7B diffusion language model, probe guidance consistently improves multiple choice question answering benchmarks.

3. 3

   A weak model used for autoguidance must come from a low-entropy region of training to ensure reliable performance.