How I Study AI - Learn AI Papers & Lectures the Easy Way

T3D: Few-Step Diffusion Language Models via Trajectory Self-Distillation with Direct Discriminative Optimization

Intermediate

Tunyu Zhang, Xinxi Zhang et al.Feb 12arXiv

This paper shows how to make diffusion language models write high‑quality text in just a few steps, which makes them much faster.

#diffusion language models#few-step decoding#trajectory self-distillation

FourierSampler: Unlocking Non-Autoregressive Potential in Diffusion Language Models via Frequency-Guided Generation

Intermediate

Siyang He, Qiqi Wang et al.Jan 30arXiv

Diffusion language models (dLLMs) can write text in any order, but common decoding methods still prefer left-to-right, which wastes their superpower.

#diffusion language models#non-autoregressive generation#frequency-domain analysis

Residual Context Diffusion Language Models

Intermediate

Yuezhou Hu, Harman Singh et al.Jan 30arXiv

Diffusion language models (dLLMs) generate several tokens at once but usually throw away lots of helpful clues each step—RCD keeps and reuses those clues.

#diffusion language models#residual context diffusion#soft tokens

Papers3

T3D: Few-Step Diffusion Language Models via Trajectory Self-Distillation with Direct Discriminative Optimization

FourierSampler: Unlocking Non-Autoregressive Potential in Diffusion Language Models via Frequency-Guided Generation

Residual Context Diffusion Language Models