Metraipolis › archive › dc_6c6feefe8aa6
Length-Adaptive Decoding for Masked Diffusion Machine Translation — arXiv:2608.22274
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.22274
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-18T07:44:44.950Z
- SHA-256 of the archived text
2cb3bbc5c9d5cf2e009232a4cd157558ab7909602d8c95ba9a02cfe199310951- Stable URL of this page
https://metraipolis.com/documento/dc_6c6feefe8aa6
How to cite
Length-Adaptive Decoding for Masked Diffusion Machine Translation — arXiv:2608.22274 — arXiv (en), https://arxiv.org/abs/2608.22274 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6c6feefe8aa6 (Metraipolis archive dc_6c6feefe8aa6, sha256 2cb3bbc5c9d5cf2e009232a4cd157558ab7909602d8c95ba9a02cfe199310951)
BibTeX
@misc{dc6c6feefe8aa6,
title = {Length-Adaptive Decoding for Masked Diffusion Machine Translation — arXiv:2608.22274},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.22274},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6c6feefe8aa6; sha256 2cb3bbc5c9d5cf2e009232a4cd157558ab7909602d8c95ba9a02cfe199310951},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.22274. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Machine translation tests masked diffusion language models (dLLMs) because every source token must be rendered faithfully, while fixed canvas decoding must choose target length before denoising. Existing masked diffusion decoding work mainly studies token unmasking order, leaving this length decision under-explored despite its direct effect on coverage and redundancy. We introduce Entropy-Valley (EV), a training-free length selector that scores candidate target canvases by mean predictive entropy from all-mask forward passes and selects the canvas the backbone is most prepared to fill. Relative to a baseline using training corpus length statistics, EV recovers 64.9%, 65.3%, and 33.0% of the COMET-22 gain from reference target lengths on En$\to$Zh, Zh$\to$En, and En$\to$De. Our diagnostics show that denoising-friendly lengths need not match reference lengths. Evaluation by three translation experts supports the En$\leftrightarrow$Zh adequacy gains, with stronger evidence on Zh$\to$En. Compared with a LLaMA-3-8B autoregressive (AR) model trained on the same fine-tuning data, the EV system ties on En$\to$Zh and leads on Zh$\to$En; an oracle-length diagnostic further shows that, in this masked diffusion MT setting, deciding which tokens to reveal first matters less than how the target length is supplied. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-25 in cs.AI). Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2608.22274 — 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 2026-10-18T07:37:52.866Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).