Metraipolis › archive › dc_cf9d5e2da405

DConstrain: Compiling Context-Parameterized Transitions for Diffusion Language Model Constrained Decoding — arXiv:2610.30920

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.30920
License (as archived)
CC0-1.0
Archived
2026-10-25T21:34:40.903Z
SHA-256 of the archived text
e52d66a5e95c6fc55dd204c92440609fc0929a0da0b3e47c8302586f559cf223
Stable URL of this page
https://metraipolis.com/documento/dc_cf9d5e2da405

How to cite

DConstrain: Compiling Context-Parameterized Transitions for Diffusion Language Model Constrained Decoding — arXiv:2610.30920 — arXiv (en), https://arxiv.org/abs/2610.30920 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cf9d5e2da405 (Metraipolis archive dc_cf9d5e2da405, sha256 e52d66a5e95c6fc55dd204c92440609fc0929a0da0b3e47c8302586f559cf223)
BibTeX
@misc{dccf9d5e2da405,
  title = {DConstrain: Compiling Context-Parameterized Transitions for Diffusion Language Model Constrained Decoding — arXiv:2610.30920},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.30920},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cf9d5e2da405; sha256 e52d66a5e95c6fc55dd204c92440609fc0929a0da0b3e47c8302586f559cf223},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.30920. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Diffusion language models (DLMs) decode multiple tokens in parallel, so individually valid tokens can be incompatible together under a schema. Exact joint inference over a grammar graph enforces compatibility, but it requires token-labeled successors across many hypothetical parser configurations. Encoding stacks and field histories in compiled states can cause exponential growth, which forces prior exact decoders to fix property orders and exclude valid outputs. We present DConstrain, which compiles schemas into context-parameterized transition programs. Cold compilation specializes token consumption to local grammar states while retaining context-dependent stack operations, environment guards, and caller-dependent deferred parsing symbolically. Tracking field histories as runtime values supports arbitrary property order without enumerating field subsets offline. At runtime, warm compilation binds reached configurations to these programs to construct exact successors, overlapping with the model forward pass because graph construction requires no model scores. Furthermore, token-effect grouping merges tokens with identical compiled effects before runtime contexts are bound, so groups are reused across stacks and environments even when concrete successors differ. Objective-specific score aggregation and token restoration provably preserve both the Viterbi optimum and the token-labeled path distribution of ungrouped inference. Across three benchmarks and two DLM families, DConstrain compiles 100% of BFCL schemas under arbitrary property orders, whereas an optimized finite-state baseline compiles only 71.30-85.81%. Under arbitrary property order, it reduces BFCL end-to-end decoding latency by 6.65-46.26 times with 100% schema validity and raises DiffusionGemma's BFCL accuracy from 37.27% under fixed order to 81.56%.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.30920 — 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 2026-10-25T21:25:18.123Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).