Improving Few-Step Language Flows with Untied Self-Conditioning

Flow-matching language models refine all token positions in parallel and can trade sampling steps for latency, yet generation quality still degrades sharply with few sampling steps. We trace a source of this degradation to how self-conditioning reuses predictions during sampling. Training constructs the conditioning input from the current noisy state. During sampling, self-conditioning ties two distinct uses to the same clean prediction: advancing the latent through the solver, and conditioning the next model evaluation. A correction that improves generation when supplied to the solver can become harmful when also used for self-conditioning. We introduce Untied Self-Conditioning, a training-free method that unties these two uses by adapting a shared prediction to their different requirements, using only one model evaluation per sampling step. The past prediction already contributes to the latent, we therefore scale down its conditioning components along directions shared by the model's input projections. For the solver update, we instead use prediction history to estimate how the clean prediction changes during a sampling step and scale the resulting correction using empirical trajectory statistics. Across several pretrained language flow baselines, our method consistently improves generation quality: with only 8 function evaluations (NFE), it reduces generative perplexity on OpenWebText from $531$ to $61$ ($8.7\times$) for LangFlow, and from $71$ to $44$ for ELF-B. Under an adapted Arena-Hard-Auto v2 evaluation on LangFlow trained on OpenWebText, our generations win $96.3\%$ of non-tied comparisons with the official sampler at the same NFE. Improvements hold from 8 to 256 sampling steps across all tested models and datasets, suggesting that our method offers a promising direction for efficient sampling in continuous diffusion language models.