Guiding Language Models of Code with Global Context using Monitors paper page: huggingface.co/papers/2306.10… Language models of code (LMs) work well when the surrounding code in the vicinity of generation provides sufficient context. This is not true when it becomes necessary to use
MGD consistently improves compilation rate and ground truth match across LM sizes. Interestingly, smaller LMs with MGD can surpass larger LMs: SantaCoder-1.1B achieves better compilation rate and next-identifier match than the much larger text-davinci-003📈
MGD complements existing code generation techniques, such as retrieval-augmented, static analysis-based prompting, architecture changes, and fill-in-the-middle decoding (see sections 4.2 & 4.3). Plus, no LM fine-tuning is required!🌟
To evaluate MGD, We curate two datasets: PragmaticCode: Real-world open-source projects, complete with development environments & dependencies. DotPrompts: Method-level completion tasks, derived from prompts in PragmaticCode📊 Code & dataset at: aka.ms/monitors4codeg…
Don't miss what's happening
People on X are the first to know.
