Reproduce “LiveMACE: Process-Aware Evaluation of LLM Agent Capabilities in Evolving Markets” 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
Jun Zhao, Leiming Fu, Yanbo Wen, Yiding Wang, Xuantong Liu, Yang Shu, Yuyang Lu, Xuanran Xing and 4 more
LiveMACEBench evaluates financial agents in persistent paper-trading accounts using live market data. It compares returns across matched configurations with decision traces that show whether tools, memory, rules, and specialist agents were used. In tool-use comparisons, return rankings reversed 21 times versus zero for Tool Call Score across 180 comparisons. This measures one approximately 30-day trajectory, not real-money performance.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.