Automated Discovery Has No Universally Superior Harness

2026-07-20

AI Summary

Although existing automated discovery systems are widely adopted as general-purpose frameworks, their performance heavily depends on component configurations and stochasticity, lacking consistent superiority across tasks. This work systematically dissects frameworks such as OpenEvolve and TTT-Discover, evaluating 30 budget-matched component combinations across 12 model–problem pairs through 3.1 million large language model inferences. The study reveals no universally optimal framework and instead proposes treating framework selection as a hyperparameter. An adaptive resource allocation strategy based on early-stage performance is introduced. Experiments demonstrate that simple alternatives often outperform complex frameworks, and the proposed adaptive strategy significantly surpasses both fixed frameworks and static ensembles, substantially improving discovery efficiency under limited computational budgets.

Abstract

Autonomous discovery systems such as OpenEvolve and TTT-Discover are often used as general-purpose harnesses. However, in practice these are composite systems combining several design choices about archives, parent selection, exploration, and budget allocation into a single recipe. Because discovery runs are expensive and inherently stochastic, existing harnesses are often compared using too few independent trials to distinguish key methodological improvements from run-to-run variance. We systematically decompose OpenEvolve-style evolutionary search and the TTT-Discover search harness into its constituent components and systematically evaluate 30 budget-matched harnesses across 12 model-problem pairs using more than 3.1 million LLM rollouts and repeated-trial statistical analysis. Our results show that discovery harnesses have a generalization problem: No fixed harness is reliably superior across the evaluated model-problem pairs, and variants of OpenEvolve generally underperform simpler alternatives. Thus, harness choice is better viewed as a hyperparameter rather than as a universal recipe, and should be tailored to the specific problem and underlying model. We also find that early discovery progress predicts final performance, and use this property to present a budget-matched adaptive-allocation experiment that starts multiple harnesses, prunes weak partial runs, and reallocates compute to stronger survivors, outperforming both commitment to a randomly sampled fixed harness and a non-adaptive harness ensemble. Together, these results motivate shifting from fixed harness selection to online adaptation guided by early performance. We release all run pools including baseline null distributions for every model-problem pair as reusable statistical infrastructure against for future harness proposals.

Problem

• automated discovery
• discovery harness
• generalization
• hyperparameter selection
• stochastic optimization

Innovation

• adaptive allocation
• discovery harness
• evolutionary search
• large language models
• hyperparameter optimization