Metraipolis › archive › dc_194fe214e40e
A Review of Small Language Models for Code and Domain-Specific Language Artifact Generation (2026) · DOI 10.20944/preprints202606.1095.v2
- Source
- Metraipolis archive
- Original
- https://doi.org/10.20944/preprints202606.1095.v2
- License (as archived)
- CC-BY-4.0
- Archived
- 2026-10-08T01:18:03.184Z
- SHA-256 of the archived text
e6582b41cedee094c0ce80d0b08f6dd60923e6f85b910a6b532feace0f9065ee- Stable URL of this page
https://metraipolis.com/documento/dc_194fe214e40e
How to cite
A Review of Small Language Models for Code and Domain-Specific Language Artifact Generation (2026) · DOI 10.20944/preprints202606.1095.v2 — Metraipolis archive, https://doi.org/10.20944/preprints202606.1095.v2 (CC-BY-4.0); archived copy: https://metraipolis.com/documento/dc_194fe214e40e (Metraipolis archive dc_194fe214e40e, sha256 e6582b41cedee094c0ce80d0b08f6dd60923e6f85b910a6b532feace0f9065ee)
BibTeX
@misc{dc194fe214e40e,
title = {A Review of Small Language Models for Code and Domain-Specific Language Artifact Generation (2026) · DOI 10.20944/preprints202606.1095.v2},
howpublished = {Metraipolis archive},
url = {https://doi.org/10.20944/preprints202606.1095.v2},
note = {License CC-BY-4.0; archived copy https://metraipolis.com/documento/dc_194fe214e40e; sha256 e6582b41cedee094c0ce80d0b08f6dd60923e6f85b910a6b532feace0f9065ee},
year = {2026}
}Archived text
Reproduced under CC-BY-4.0. Attribution: Metraipolis archive, https://doi.org/10.20944/preprints202606.1095.v2. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Articolo recente con licenza aperta: scheda dai metadati Crossref (posted-content). Titolo: A Review of Small Language Models for Code and Domain-Specific Language Artifact Generation Autori: Andrey V. Kozachok Pubblicato: 2026-08-19 · editore: MDPI AG DOI: https://doi.org/10.20944/preprints202606.1095.v2 Licenza dell'articolo: CC-BY-4.0 Abstract (testo degli autori, CC-BY-4.0): Small Language Models (SLMs, up to 7B parameters) offer a resource-efficient alternative to large models for code generation tasks, yet existing studies of their application to domain-specific languages (DSLs) use heterogeneous task formulations and evaluation protocols. This paper reviews SLMs for DSL artifact generation, including configurations, security policies, and infrastructure templates. Based on an analysis of the literature, we propose a classification of DSL artifact generation tasks according to the composition of the required context model. Classes C1–C3 represent a progression from local structural constraints to schema and inter-element relationships and then to dependencies on semantics and the execution environment. For these classes, we formulate hypotheses about model adaptation strategies. We systematize adaptation methods, including LoRA/QLoRA, distillation, grammar-constrained decoding, and RAG, and highlight automatic formal verification as a key advantage of DSLs for reliable generation in critical infrastructure settings. Fonte: https://api.crossref.org/works/10.20944%2Fpreprints202606.1095.v2 — metadati Crossref, consultati il 2026-10-07. Interrogazione: «small language models».