Fetching from the wire…
01
02
03
04
05
06
07
08
Research2026-06-18 · source-backed
Two papers land the same week: DreamReasoner-8B uses block-size curriculum learning to scale parallel block-wise denoising for long chain-of-thought (arXiv 2606.19257), and Diffusion-Proof applies diffusion-style generation to formal theorem proving (arXiv 2606.19315). The case that diffusion LMs can compete with autoregressive models on structured reasoning, while decoding in parallel, is getting harder to wave off as a curiosity.
Each link below shares sources, entities, or timing with this story.
Shared entity: LMs / Same source domain / Shared topic / What happened next
Both cover LMs; reported by the same outlet (arxiv.org); overlapping topics (denoising, diffusion).
Shared entity: LMs / Shared topic / What happened next
Both cover LMs; overlapping topics (autoregressive, decoding, diffusion, parallel); picks up the LMs thread on 2026-06-22.
Shared entity: Diffusion / Shared topic / Earlier coverage
Both cover Diffusion; overlapping topics (decoding, diffusion, model, parallel); earlier Diffusion coverage from 2026-06-11.
Shared entity: Diffusion / Same source domain / Shared topic / Earlier coverage
Both cover Diffusion; reported by the same outlet (arxiv.org); overlapping topics (diffusion, model).
Same source domain / Shared topic / Tension
Reported by the same outlet (arxiv.org); overlapping topics (autoregressive, decoding, diffusion, model); pushes against this story (but).
Reported by the same outlet (arxiv.org); overlapping topics (decoding, model, reasoning); pushes against this story (but).
Shared entity: Proof / Same source domain / What happened next
Both cover Proof; reported by the same outlet (arxiv.org); picks up the Proof thread on 2026-08-18.
Shared entity: LMs / Shared topic / What happened next
Both cover LMs; overlapping topics (diffusion, model); picks up the LMs thread on 2026-07-28.