Discrete diffusion or flow models could enable faster and more controllable sequence generation than autoregressive models. We show that na\"ive linear flow matching on the simplex is insufficient toward this goal since it suffers from discontinuities in the training target and further pathologies. To overcome this, we develop Dirichlet flow matching on the simplex based on mixtures of Dirichlet distributions as probability paths. In this framework, we derive a connection between the mixtures' scores and the flow's vector field that allows for classifier and classifier-free guidance. Further, we provide distilled Dirichlet flow matching, which enables one-step sequence generation with minimal performance hits, resulting in $O(L)$ speedups compared to autoregressive models. On complex DNA sequence generation tasks, we demonstrate superior performance compared to all baselines in distributional metrics and in achieving desired design targets for generated sequences. Finally, we show that our classifier-free guidance approach improves unconditional generation and is effective for generating DNA that satisfies design targets. Code is available at https://github.com/HannesStark/dirichlet-flow-matching.

In this presentation, Kathrin Melcher, who works as a data scientist at KNIME, will give an overview of KNIME Software, including the open-source tool KNIME Analytics Platform for creating data science applications and services and also the different deployment options you have when using KNIME Server. While the structure is often similar--data collection, data transformation, model training, deployment--each project required its own special trick, whether this was a change in perspective or a particular technique to deal with the special case and business questions. You'll learn about demand prediction in energy, anomaly detection in IoT, risk assessment in finance, the most common applications in customer intelligence, social media analysis, topic detection, sentiment analysis, fraud detection, bots, recommendation engines, and more. Join us to learn what's possible in data science. She holds a Master's Degree in Mathematics from the University of Konstanz, Germany.