Reproduce “The Era by Eon Benchmark: A Generated Enterprise Estate with Exact Ground Truth for Benchmarking LLM 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
Benjamin Gruenbaum, Doron Porat, Assaf Natanzon, Roy Zavida, Chen Dinachi, Or Itzahary
The authors present a benchmark for evaluating enterprise agents by generating one fictional company whose records appear across product simulators and company-specific databases. Seeded generation keeps systems linked, and answer keys are computed from the finished records. Across 23 companies, realism-scorecard mean rose from 61.8 to 97.0; on one estate, nine models’ accuracy estimates ranged from 42.4% to 76.8%, not a general ranking.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.