Metraipolis › archive › dc_f3c8381cf9fc

ReaDy: Recurrent States as Dynamic Context for Speculative Drafting in Hybrid Linear Attention Large Language Models — arXiv:2610.20503

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.20503
License (as archived)
CC0-1.0
Archived
2026-10-15T20:35:54.544Z
SHA-256 of the archived text
5705d433ea391b1dbbfdd220b5be78f0c0e7cf1daea1b57a7dfec8b2da4d8859
Stable URL of this page
https://metraipolis.com/documento/dc_f3c8381cf9fc

How to cite

ReaDy: Recurrent States as Dynamic Context for Speculative Drafting in Hybrid Linear Attention Large Language Models — arXiv:2610.20503 — arXiv (en), https://arxiv.org/abs/2610.20503 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f3c8381cf9fc (Metraipolis archive dc_f3c8381cf9fc, sha256 5705d433ea391b1dbbfdd220b5be78f0c0e7cf1daea1b57a7dfec8b2da4d8859)
BibTeX
@misc{dcf3c8381cf9fc,
  title = {ReaDy: Recurrent States as Dynamic Context for Speculative Drafting in Hybrid Linear Attention Large Language Models — arXiv:2610.20503},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.20503},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f3c8381cf9fc; sha256 5705d433ea391b1dbbfdd220b5be78f0c0e7cf1daea1b57a7dfec8b2da4d8859},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.20503. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Speculative decoding accelerates autoregressive generation by drafting multiple tokens per step, yet feature-based drafters rely on target hidden features whose contextual emphasis may not align with the needs of subsequent draft positions. Hybrid linear attention (HLA) models maintain compact recurrent states that accumulate historical associations, offering an efficient source of complementary context for speculative drafting. We introduce ReaDy, the first speculative decoding method, to the best of our knowledge, to leverage HLA recurrent states for position-specific context retrieval during drafting. ReaDy equips each draft position with a learned query to retrieve supplementary context directly from the target's committed recurrent states. A lightweight state reader integrates this context into draft representations through residual updates, while low-rank state reads and fused operators keep additional computation and I/O overhead low. Building on DSpark's parallel backbone and sequential prediction head, ReaDy employs a lightweight two-stage adaptation strategy, with reader capacity, state sources, and residual placement examined through ablation studies. Across seven benchmarks spanning mathematics, code generation, and dialogue, ReaDy achieves an average $3.16\times$ decoding speedup over autoregressive decoding with Qwen3.8-27B on a single NVIDIA B200 GPU. It outperforms DSpark on every benchmark, improving throughput by 7.97% and acceptance length by 8.50% on average. Our code is available at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.20503 — 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 2026-10-15T19:36:18.272Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).