DELA: Learning Controllable Edits for Diffusion Language Models
Trains a small controller to insert, delete or remask parts of a text draft, while a frozen diffusion language model fills the gaps. Tests whether these targeted edits improve mathematical answers and code; intermediate reasoning remains explicit text.
First public: · Added to this list: October 26, 2026
Explore this approach
Alternative generation methods — Change how intermediate steps are generated, revised, or searched—for example, by refining several steps in parallel. Some produce written chains of thought.