dvoting-diffusion-consensus

【开发与工程】Accelerate test-time scaling for diffusion language models by identifying inconsistent tokens, selectively remask and regenerate only uncertain tokens, and aggregate across samples via voting. Achieve 5.5-22× speedup over standard iterative sampling with 6-8% accuracy gains on reasoning tasks.

独立 GitHub 快照 · 自动分类 · 以原作者文档为准