Reproduce “OpenSaC: A Python Library for Search as Code in Deep Research Agents” 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
Qi Liu, Jiaxin Mao, Fengbin Zhu, Tat-Seng Chua, Ji-Rong Wen
OpenSaC presents an open Python library and companion Search as Code skill for coding agents to compose research searches and processing steps into executable programs. The agent can batch requests, organize returned material, and send selected evidence back for decisions. In two benchmarks, this configuration ran faster on both, with higher quality on BrowseComp but slightly lower quality on WideSearch.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.