Metraipolis › archive › dc_4bcb7949867c
DA-DLM: Explicitly Modeling Token Dependencies in Diffusion Language Models — arXiv:2609.15070
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.15070
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-15T23:15:50.778Z
- SHA-256 of the archived text
47e29f54059912466c4e2766d1332d03e939ce74beca693e43ebe30c4da5f030- Stable URL of this page
https://metraipolis.com/documento/dc_4bcb7949867c
How to cite
DA-DLM: Explicitly Modeling Token Dependencies in Diffusion Language Models — arXiv:2609.15070 — arXiv (en), https://arxiv.org/abs/2609.15070 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4bcb7949867c (Metraipolis archive dc_4bcb7949867c, sha256 47e29f54059912466c4e2766d1332d03e939ce74beca693e43ebe30c4da5f030)
BibTeX
@misc{dc4bcb7949867c,
title = {DA-DLM: Explicitly Modeling Token Dependencies in Diffusion Language Models — arXiv:2609.15070},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.15070},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4bcb7949867c; sha256 47e29f54059912466c4e2766d1332d03e939ce74beca693e43ebe30c4da5f030},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.15070. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Diffusion Language Models (DLMs) generate text by iteratively denoising a masked sequence, independently predicting multiple tokens at each step. This conditional independence discards inter-token dependencies and degrades coherence-an issue that parallels the multi-modality problem in Non-Autoregressive Translation (NAT). Drawing on the Directed Acyclic Transformer (DAT), which tackles this problem in NAT via a Directed Acyclic Graph (DAG), we propose DA-DLM, a model that adapts DAG-based dependency modeling to DLMs' iterative setting through a position-oriented DAG design. The position-oriented DAG binds node groups to fixed output positions so that tokens fixed in earlier steps anchor neighboring predictions via learned transitions, and evolves with denoising to focus on remaining uncertainty as anchors accumulate. On language modeling, open-ended generation, and summarization, DA-DLM consistently outperforms Block Diffusion, especially under fewer denoising steps, and matches autoregressive models while preserving the parallel generation advantage. Our code is publicly available at this https URL . Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-15 in cs.CL). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2609.15070 — 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 2026-10-15T22:51:09.929Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).