Sampling from Flow Language Models via Marginal-Conditioned Bridges
Researchers have introduced a new sampling method for Flow Language Models (FLMs) called marginal-conditioned bridges. This technique adapts continuous flow matching for token sequences, addressing limitations in standard diffusion model samplers. The proposed method samples endpoints from FLM token marginals and then uses an analytic Ornstein-Uhlenbeck bridge, offering improved quality-diversity tradeoffs and principled control over decoding. AI
IMPACT Introduces a novel sampling technique that enhances the quality-diversity balance in Flow Language Models.