Search papers, labs, and topics across Lattice.
This paper presents a novel literate programming environment that integrates language-aware machine agents with a grammar for executable program essays and a first-class name-graph. By enabling the co-location of code with relevant natural language and structured data, the environment enhances the context utilization of Large Language Models (LLMs) and provides coding agents with tools akin to those found in human-facing Integrated Development Environments (IDEs). The implementation includes bindings to three programming languages and demonstrates improved interaction between prose and executable artifacts through practical examples.
LLM coding agents can now leverage a literate programming environment that enhances context utilization and mimics human IDE capabilities.
This paper introduces an environment for constructing literate programs in concert with language-aware machine agents. This environment includes a grammar for executable program essays, a parser that treats names as first-class objects, an internal name-graph which relates prose, names and executable artifacts, and a binding mechanism for existing languages and testing toolsets. This supports co-location of code with its most relevant natural language and structured data context, making better use of Large Language Model (LLM) context windows. It also provides LLM coding agents with a toolset more analogous to the symbol-aware search and usage information available in human programmer-facing Integrated Development Environments (IDEs). We describe a working implementation with bindings to three established programming languages, and several example programs.
