Reproduce “Concurrency without Model Changes: Future-based Asynchronous Function Calling for LLMs” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Guangyu Feng, Huanzhi Mao, Prabal Dutta, Joseph E. Gonzalez
AsyncFC reduces end-to-end task latency in tool-using language model workflows by returning future placeholders while tools run and letting decoding continue. A dependency-aware scheduler controls execution using resource annotations. On 300-task SWE-bench Lite evaluation, it reports 1.40× speedup over SWE-agent, with issue-resolution rates of 79.3% versus 79.0%; latency significance is against +Par. FC, and accuracy equivalence is not established.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.