# Loreley: Repository-Scale Program Evolution with Quality-Diversity Search

arXiv: https://arxiv.org/abs/2608.19703 · HTML: https://mlsanity.com/paper/2608.19703/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 31 references that are in the map.
- First posted on arXiv 20 Aug 2026.

## Builds on (most cited first, 18 of 18)

- [AlphaEvolve: A coding agent for scientific and algorithmic discovery](https://mlsanity.com/paper/2506.13131/) (2025)
- [Illuminating search spaces by mapping elites](https://mlsanity.com/paper/1504.04909/) (2015)
- [Evolution through Large Models](https://mlsanity.com/paper/2206.08896/) (2022)
- [Autonomous skill discovery with Quality-Diversity and Unsupervised Descriptors](https://mlsanity.com/paper/1905.11874/) (2019)
- [CodeEvolve: an open source evolutionary coding agent for algorithmic discovery and optimization](https://mlsanity.com/paper/2510.14150/) (2025)
- [RHO: Your Coding Agent is Secretly a Roboticist](https://mlsanity.com/paper/2606.16458/) (2026)
- [Multi-Objective Quality Diversity Optimization](https://mlsanity.com/paper/2202.03057/) (2022)
- [Autonomous Code Evolution Meets NP-Completeness](https://mlsanity.com/paper/2509.07367/) (2025)
- [Quality-Diversity Algorithms Can Provably Be Helpful for Optimization](https://mlsanity.com/paper/2401.10539/) (2024)
- [EvoGit: Decentralized Code Evolution via Git-Based Multi-Agent Collaboration](https://mlsanity.com/paper/2506.02049/) (2025)
- [Simple Baselines are Competitive with Code Evolution](https://mlsanity.com/paper/2602.16805/) (2026)
- [FML-bench: A Controlled Study of AI Research Agent Strategies from the Perspective of Search Dynamics](https://mlsanity.com/paper/2605.17373/) (2026)
- [Autonomous Evolution of EDA Tools: Multi-Agent Self-Evolved ABC](https://mlsanity.com/paper/2604.15082/) (2026)
- [Effective Harness Engineering for Algorithm Discovery with Coding Agents](https://mlsanity.com/paper/2605.15221/) (2026)
- [GEAR: Genetic AutoResearch for Agentic Code Evolution](https://mlsanity.com/paper/2605.13874/) (2026)
- [What Do Evolutionary Coding Agents Evolve?](https://mlsanity.com/paper/2605.20086/) (2026)
- [Heuresis: Search Strategies for Autonomous AI Research Agents Across Quality, Diversity and Novelty](https://mlsanity.com/paper/2606.25198/) (2026)
- [QDEvo: A Multi-Objective Quality-Diversity Framework for Automated Heuristic Design](https://mlsanity.com/paper/2607.11916/) (2026)

## Abstract (arXiv)

Sequential agent search accumulates changes from its current champion but discards alternative branches; independent proposals preserve breadth but restart from the root. Loreley instead retains complete repository states in a Quality-Diversity (QD) archive and samples them as parents or supplies them as context for later edits. Candidates are Git commits produced in isolated worktrees and judged by a project-supplied evaluator. We compare configured Loreley QD, sequential champion editing, and independent root proposals in a matched Zstandard experiment: seven paired blocks and 48 physical candidate jobs per policy and block (1,008 total), with root-only initialization and each policy's native concurrency. Validation selected a winner at each budget checkpoint; an agent-hidden holdout measured the fixed candidate. At 48 jobs, QD was 0.135% below Sequential Champion (95% BCa interval for the paired effect: -0.556% to +0.161%) and 0.320% above Independent Root (-0.082% to +0.686%). Neither contrast established a QD advantage; Sequential had the highest observed 48-job mean and median. Archive retention and later sampling did occur. Four of seven final QD winners had a non-incumbent state in their primary-parent ancestry under a retrospective one-incumbent rule applied only to the observed QD stream. Including inspiration edges raised the count to six, without showing that supplied context caused an edit. Three earlier capability campaigns produced generation-4, multi-file improvements in two Python libraries and a separate Zstandard revision. Loreley engaged the intended stepping-stone mechanism, but the controlled experiment did not show an endpoint benefit at 48 jobs.