Metraipolis › archive › dc_77fe744400f2

Unlocking Lossless Speedups in LLMs via Discrete Diffusion — arXiv:2609.04010

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.04010
License (as archived)
CC0-1.0
Archived
2026-10-17T02:08:26.466Z
SHA-256 of the archived text
5ba07e3291328756bc1ebb837686644116eef7dae537212773bd3a9a4c3b2786
Stable URL of this page
https://metraipolis.com/documento/dc_77fe744400f2

How to cite

Unlocking Lossless Speedups in LLMs via Discrete Diffusion — arXiv:2609.04010 — arXiv (en), https://arxiv.org/abs/2609.04010 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_77fe744400f2 (Metraipolis archive dc_77fe744400f2, sha256 5ba07e3291328756bc1ebb837686644116eef7dae537212773bd3a9a4c3b2786)
BibTeX
@misc{dc77fe744400f2,
  title = {Unlocking Lossless Speedups in LLMs via Discrete Diffusion — arXiv:2609.04010},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.04010},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_77fe744400f2; sha256 5ba07e3291328756bc1ebb837686644116eef7dae537212773bd3a9a4c3b2786},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.04010. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Large Language Models (LLMs) owe much of their success to next-token prediction (NTP), but their autoregressive (AR) structure requires slow, sequential token generation. To overcome this bottleneck, we introduce diffusion-augmented LLMs, a new class of models that defines an AR model distribution while using diffusion to draw multiple tokens in parallel from that distribution. We decouple the parameters of these models into two sets: AR weights, trained using the standard NTP objective, and lightweight diffusion weights, trained to generate multiple tokens simultaneously. The diffusion weights are learned through a simple Diffusion Distillation phase that adds negligible overhead to existing LLM training pipelines. We also introduce $\Psi$-Spec, a family of samplers that enables lossless acceleration and inference-time scaling at a fixed context length. Unlike speculative decoding, our method requires no separate draft model. Unlike diffusion LLMs (d-LLMs), it accelerates generation without sacrificing the quality of the underlying AR model. The resulting models, called Uno, can be trained from scratch or built by augmenting existing open-weight AR LLMs. Uno achieves higher throughput than leading speculative-decoding methods at every evaluated batch size and delivers up to $3\times$ speedups over the base AR model, including at the largest batch size supported by the device. Notably, our 8B Uno model outperforms the leading open d-LLM, the 26B DiffusionGemma, and the proprietary Mercury 2 across all evaluated benchmarks in agentic tool use, coding, and long-context reasoning. We release code and checkpoints at: this https URL

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-04 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.04010 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T01:47:25.582Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-04?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).