Reproduce “From Existing Methods to New Research Ideas: Two Perspectives on Recursive Self-Improvement” 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
Leijun Zhou, Yanke Yu, Zhihao Liu, Qu Xiang, Tiancheng Hu, Yiru Dong, Yihan Wu, Yangxi Xiang and 5 more
The authors present a survey and framework for comparing recursive self-improvement (RSI) systems. They organize design choices into nine dimensions covering what changes, who proposes updates, how versions are organized, and how feedback guides retention, then connect existing methods to learning mechanisms to suggest research directions. A logging example illustrates delayed value; the paper reports no benchmark or experimental validation.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.