Reproduce “Absorbing State Phase Transitions in Multi-Agent Search” 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
Wenwen Zheng, Yuzhe Yang, Helen Qu, Xin Eric Wang, Haewon Jeong
The authors develop a framework for how shared evidence affects multi-agent search, defining critical degree as the communication needed to stop incorrect lineages from growing. Theory predicts greater pruning when components recur across search paths. Deterministic tests track theory closely (TCAS concordance 0.97), while Qwen3.5-4B’s synthetic-task threshold rises from 6.55 to 12.82 under agent-chosen communication; this measures lineage growth, not task success.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.