arXiv:2605.20199v1 Announce Type: cross
Abstract: We present FlowLM, a flow matching language model transformed from pre-trained diffusion language models via efficient fine-tuning. By re-aligning the curved sampling trajectories of diffusion models into straight-line flows, FlowLM enables high quality few-step generation that rivals or even outperforms the quality of 2,000-step diffusion sampling with very few training epochs. Remarkably, finetuned FlowLM reaches performance saturation with only half as many training epochs as training from scratch, both approaches greatly outperforming the original diffusion model, thereby validating our method. Furthermore, we validate a more effective training objective for flow matching: predicting clean data to consistently guide the sampling process towards the true data distribution. Empirical results demonstrate that our approach is highly effective for high-quality, few-step text generation.
Patient and clinician perceptions, expectations, and usability of ankle exoskeletons for daily living: a mixed-methods survey study
Ankle exoskeletons offer promising support for individuals with chronic foot drop, yet user and clinician perspectives on their use in daily living remain underexplored. Related