REVIEW 6 cited by
DINGO: Constrained Inference for Diffusion LLMs
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
DINGO: Constrained Inference for Diffusion LLMs
read the original abstract
Diffusion LLMs have emerged as a promising alternative to conventional autoregressive LLMs, offering significant potential for improved runtime efficiency. However, existing diffusion models lack the ability to provably enforce user-specified formal constraints, such as regular expressions, which makes them unreliable for tasks that require structured outputs, such as fixed-schema JSON generation. Unlike autoregressive models that generate tokens sequentially, diffusion LLMs predict a block of tokens in parallel. This parallelism makes traditional constrained decoding algorithms, which are designed for sequential token prediction, ineffective at preserving the true output distribution. To address this limitation, we propose DINGO, a dynamic programming-based constrained decoding strategy that is both efficient and provably distribution-preserving. DINGO enables sampling of output strings with the highest probability under the model's predicted distribution, while strictly satisfying any user-specified regular expression. On standard symbolic math and JSON generation benchmarks, DINGO achieves up to a 68 percentage point improvement over unconstrained inference
Forward citations
Cited by 6 Pith papers
-
Imprompt: A Language Framework for Prompt Programming
Prompt programs can be typed like programs and compiled in different ways; Imprompt formalizes two such compilers and tests them on tax-law reasoning and data-labeling tasks.
-
Continuous Diffusion Models Can Obey Formal Syntax
Training-free gradient guidance, based on the exact regex-acceptance probability under the decoder's unigram distribution, lets continuous diffusion language models satisfy regular-expression constraints.
-
Constraint Tax in Open-Weight LLMs: An Empirical Study of Tool Calling Suppression Under Structured Output Constraints
Open-weight LLMs exhibit tool suppression under joint tool-calling and JSON-schema constraints due to grammar token masking; a two-pass inference method restores tool use.
-
RefineStat: Efficient Exploration for Probabilistic Program Synthesis
RefineStat improves small language model performance on probabilistic program synthesis by adding semantic constraint enforcement and diagnostic-aware refinement, producing syntactically and statistically reliable cod...
-
The Hidden Cost of Structured Generation in LLMs: Draft-Conditioned Constrained Decoding
Draft-then-constrain decoding—first generate a free-form draft, then run format-constrained decoding conditioned on that draft—consistently improves structured-output accuracy across 1B–14B models.
-
Discrete Diffusion Models: A Unified Framework from Tokenization to Generation
Discrete diffusion models are re-framed as instances of a tokenization-centric, four-component design space (corruption, denoiser, objective, sampler) in a broad survey with no new experimental or theoretical results.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.