1 paper · 1 filter
Yang Li, Han Meng, Chenan Wang +1
Diffusion language models (DLMs) have shown strong potential for general natural language tasks with in-context examples. However, due to the bidirectional attention mechanism, DLM…