Skip to content
Hugging Face Daily Papers·· 2026-10-14AI score22

Self-conditioning enables deep sequential reasoning in Flow Matching

Self-Conditioning Enables Deep Sequential Reasoning in Flow Matching

AI briefing

The study finds that the trajectories of a native flow matching sampler can be simulated by a constant number of parallel model evaluations, and that increasing sampling steps cannot improve the expressive power of a fixed-depth backbone model. Self-conditioning (using the previous step's estimate as a condition) breaks this bottleneck: across multiple visual reasoning tasks, flow matching with self-conditioning achieves higher accuracy and greater reasoning depth, and performance improves as sampling steps increase; theoretically, a fixed-depth backbone can reach the expressive power of a space-bounded Turing machine.

Open original arxiv.org