Metraipolis › archive › dc_573caa95dbe3

Ripple-Pivot Search: Active Parallel Decoding for Diffusion Large Language Models — arXiv:2608.11742

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.11742
License (as archived)
CC0-1.0
Archived
2026-10-19T05:27:35.834Z
SHA-256 of the archived text
634cdfe18deca7300fc6a7f7ac8c1419fb0e0d59fed7cb40d08ddbf4f370c068
Stable URL of this page
https://metraipolis.com/documento/dc_573caa95dbe3

How to cite

Ripple-Pivot Search: Active Parallel Decoding for Diffusion Large Language Models — arXiv:2608.11742 — arXiv (en), https://arxiv.org/abs/2608.11742 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_573caa95dbe3 (Metraipolis archive dc_573caa95dbe3, sha256 634cdfe18deca7300fc6a7f7ac8c1419fb0e0d59fed7cb40d08ddbf4f370c068)
BibTeX
@misc{dc573caa95dbe3,
  title = {Ripple-Pivot Search: Active Parallel Decoding for Diffusion Large Language Models — arXiv:2608.11742},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.11742},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_573caa95dbe3; sha256 634cdfe18deca7300fc6a7f7ac8c1419fb0e0d59fed7cb40d08ddbf4f370c068},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.11742. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Diffusion Large Language Models (dLLMs) have emerged as a competitive alternative to autoregressive language models, offering the potential for substantially faster inference through parallel decoding. Existing parallel decoding schedulers typically commit positions only after they meet a per-position criterion, overlooking how early commitments may benefit subsequent decoding. We identify a ripple effect in dLLM decoding: proactively committing a mid-entropy pivot position can induce a pronounced reduction in uncertainty across the remaining masked positions. This uncertainty reduction allows subsequent steps to unmask more tokens in parallel, thereby accelerating the overall decoding process. To exploit the ripple effect, we propose Ripple-Pivot Search (RPS), a novel training-free decoding method that seeks mid-entropy positions as promising candidate pivots (where to decode), and determines their token assignment that yields the greatest downstream benefit via lookahead evaluation (what to decode). Across 3 dLLMs and 4 reasoning and code-generation benchmarks, RPS achieves 4-10$\times$ wall-clock speedup over the standard decoder while preserving generation quality, and improves accuracy over the previous lookahead baseline by up to 5.49% while delivering higher throughput in most settings. When integrated with KV caching, RPS further achieves up to 18$\times$ wall-clock speedup over the standard decoder.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-13 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2608.11742 — 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 2026-10-19T05:15:39.208Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-13?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).