---
title: "Automated Discovery Has No Universally Superior Harness"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2607.18235v1"
permalink: "https://feed.astralor.com/items/itm_05fcc563a479cb9ab9352a47"
published: "2026-07-20T17:59:37.000Z"
score: 81
channel: "ai"
tags: ["OpenEvolve","TTT-Discover","Large Language Models","Automated Discovery"]
---

# Automated Discovery Has No Universally Superior Harness

## AI commentary

> This paper is highly recommended for teams working on AutoML, agent search, and LLM-driven automation. It provides strong empirical evidence against blindly adopting existing harnesses and advocates for task-specific adaptation. The adaptive allocation strategy offers a plug-and-play improvement, and the released statistical infrastructure can accelerate future benchmarking.

## At a glance

This paper systematically decomposes the components of automated discovery harnesses like OpenEvolve and TTT-Discover, and evaluates 30 budget-matched harnesses across 12 model-problem pairs using over 3.1 million LLM rollouts with repeated-trial statistical analysis. The results show no fixed harness is universally superior, and OpenEvolve variants generally underperform simpler alternatives. The authors suggest treating harness choice as a hyperparameter rather than a universal recipe, and propose an adaptive allocation strategy based on early discovery progress. All run pools and baseline null distributions are released as reusable statistical infrastructure.

## Key points

- Systematically compared 30 discovery harnesses across 12 model-problem pairs using over 3.1 million LLM rollouts and repeated-trial statistics.
- No fixed harness is reliably superior across all settings, and OpenEvolve variants generally underperform simpler alternatives.
- Early discovery progress predicts final performance, enabling an adaptive allocation strategy that outperforms fixed harness and non-adaptive ensemble.

## Full text

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.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_05fcc563a479cb9ab9352a47