Search for a command to run...
WeDLM: Reconciling Diffusion Language Models with Standard Causal Attention for Fast Inference