Register Tokens for Bounded-State Reasoning in Diffusion Language Models
Register tokens let diffusion language models like LLaDA and Dream carry reasoning state across cleared chunks, gaining up to 19.5 points on code.
1 stories, newest first · one card per story
Register tokens let diffusion language models like LLaDA and Dream carry reasoning state across cleared chunks, gaining up to 19.5 points on code.
Today 1.6K readers · 2.5K views · 779 reads · all time 189K views