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

By Sun, Su, Zhuang +3Oct 14, 2026

Score████░░░░░░3.8

Key numbers

3.16x decoding speedup over autoregressive
7.97% and acceptance length by
8.50% on average

Caveats

Preprint; not yet peer reviewed.

VerdictWorth a reader's time today.

Read the original PDF Code

Abstract

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× 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 https://github.com/StartLuxLabs/ReaDy.

Bohan Sun, Zunhai Su, Xialie Zhuang, Jianjin Li, Ngai Wong, Chuan-Wei Kuo

The editor's rubric

Heuristic review

Leverage4 / 5: A general-purpose tool used across several fields (Adam, ResNet, LoRA, next-generation sequencing).; 24% of this desk’s rubric
Magnitude4 / 5: A qualitative jump: a capability or regime that did not exist before.; 18% of this desk’s rubric
Evidence3 / 5: Solid: multiple benchmarks or cohorts, ablations, fair baselines, released code or data.; 14% of this desk’s rubric
Novelty3 / 5: A genuinely new approach to an open problem.; 16% of this desk’s rubric
Trajectory3 / 5: A clear path to scale.; 18% of this desk’s rubric
Stakes2 / 5: Benefits a professional community (practitioners, clinicians, engineers).; 10% of this desk’s rubric

Dimension	Level	Weight	What that level means
Leverage	████░ 4	24%	A general-purpose tool used across several fields (Adam, ResNet, LoRA, next-generation sequencing).
Magnitude	████░ 4	18%	A qualitative jump: a capability or regime that did not exist before.
Evidence	███░░ 3	14%	Solid: multiple benchmarks or cohorts, ablations, fair baselines, released code or data.
Novelty	███░░ 3	16%	A genuinely new approach to an open problem.
Trajectory	███░░ 3	18%	A clear path to scale.
Stakes	██░░░ 2	10%	Benefits a professional community (practitioners, clinicians, engineers).

Editor’s rationale

Heuristic triage from title and abstract text only, not a reading of the paper. Cues found: method (we propose); breadth (many tasks); gains (x-fold, relative gain, outperforms); firsts (first); verification (multiple benchmarks, ablations, code released); scale (efficient); stakes (general AI).

How the score was computed

rank-2026-10-18

Score████░░░░░░3.8

Score = 10 × (65% × adjusted merit / 10 + 25% × attention + 10% × freshness)

Desks place stronger verdicts first, then use this score to order work within the same verdict.

Merit: 6.6 / 10; Weighted rubric, evidence-gated.
Adjusted merit: 5.3 / 10; Shrunk toward the desk prior by editor confidence (50%).
Attention: 0%; Citations, upvotes, points, mentions.
Freshness: 34%; Half-life decay since publication.

Citations0 (reference 15, via semantic-scholar, Oct 17, 2026, 13:49 UTC)
Influential citations0 (reference 3, via semantic-scholar, Oct 17, 2026, 13:49 UTC)

The record

Reviewed by heuristic-v7 on Oct 18, 2026, 13:49 UTC. Paper type: method.
Categories: cs.CL
BRIEF, No.9 in the Artificial Intelligence edition of October 18, 2026.