pith. sign in

CoRR , volume =

2 Pith papers cite this work. Polarity classification is still indexing.

2 Pith papers citing it

fields

cs.CL 2

years

2026 2

verdicts

UNVERDICTED 2

representative citing papers

Speculative Decoding Across Languages

cs.CL · 2026-05-28 · unverdicted · novelty 4.0

N-gram draft models give larger and more consistent speed-ups for multilingual speculative decoding than fine-tuned neural drafts, despite lower acceptance rates, across translation and story generation.

citing papers explorer

Showing 2 of 2 citing papers.

  • ConSA: Controllable Sparsity in Hybrid Attention via Learnable Allocation cs.CL · 2026-06-16 · unverdicted · none · ref 32

    ConSA learns FA/SWA allocation via L0 masks and augmented Lagrangian constraints, outperforming rule-based baselines on 0.6B and 1.7B models with consistent layer patterns.

  • Speculative Decoding Across Languages cs.CL · 2026-05-28 · unverdicted · none · ref 27

    N-gram draft models give larger and more consistent speed-ups for multilingual speculative decoding than fine-tuned neural drafts, despite lower acceptance rates, across translation and story generation.