Reproduce “The World Won't Wait: Evaluating and Teaching Agents Under a Live Clock” 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
Ningxin Pan, Hanyu Li, Yeqing Shen, Ning Ding, Yehui Tang, Wentao Zhang
TORSI moves slow teacher reasoning and practice between attempts, then gives reusable guidance to Jev, which acts without teacher calls during live games. In RealtimeBench, Jev’s mean test score rises from 13.51 to 24.19 after two rounds, improving across all six game types. These results use structured states and normalized peak progress, which does not necessarily indicate a win.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.