# DocLang: a markup language for LLMs

DevFeed: [DocLang: a markup language for LLMs](<https://devfeed.tech/articles/doclang-a-markup-language-for-llms-17332.md>)

Original publisher: [Read original article](<https://research.ibm.com/blog/doclang-ai-native-doc-standard>)

Author: Kim Martineau

Published: 2026-08-12T12:00:00Z

Content type: article

Language: en

Sources: [IBM Research](<https://devfeed.tech/sources/ibm-research.md>)

Topics: [XML](<https://devfeed.tech/topics/xml.md>), [Parser](<https://devfeed.tech/topics/parser.md>), [Parsing](<https://devfeed.tech/topics/parsing.md>), [Large Language Model](<https://devfeed.tech/topics/llm.md>), [Generative AI](<https://devfeed.tech/topics/generative-ai.md>), [HTML](<https://devfeed.tech/topics/html.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [algorithms](<https://devfeed.tech/tags/algorithms.md>), [generative-ai](<https://devfeed.tech/tags/generative-ai.md>), [llms](<https://devfeed.tech/tags/llms.md>), [news](<https://devfeed.tech/tags/news.md>), [parsing](<https://devfeed.tech/tags/parsing.md>), [xml](<https://devfeed.tech/tags/xml.md>)

## AI overview

IBM researchers introduce DocLang, a constrained XML sub-language designed to help large language models interpret document structure. The article explains how it represents semantics, layout, bounding boxes, and reading order, and how it may reduce token use, latency, and inference costs in AI pipelines.

## Source excerpt

The lead researcher behind IBM's popular document parser, Docling, explains why generative AI needs its own document standard.