Reproduce “HACO: Hedged Agent Computing for Reliable LLM Systems” 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

Enhan Li, Hongyang Du

HACO selects candidate deployments as a hedge set for each role request, rather than using one executor or the full pool. It ranks candidates by estimated quality, reliability, latency, and uncertainty, and uses execution traces to inform allocations. On MatplotBench, it scored 58.86 versus 42.75 for full-pool Best-of-N, with equal 36% strict success but fewer tokens and less time in simulation.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.