Search papers, labs, and topics across Lattice.
Loreley introduces a Quality-Diversity (QD) search approach that retains complete repository states, allowing for broader exploration of program evolution compared to traditional sequential champion editing and independent root proposals. In a controlled experiment with 1,008 candidate jobs, Loreley achieved performance metrics that were marginally below sequential champion editing but above independent root proposals, indicating no definitive advantage for QD in this context. Notably, several QD winners emerged from non-incumbent states, suggesting potential benefits in maintaining diverse ancestry for future edits, despite the lack of a clear endpoint advantage at the tested job count.
Quality-Diversity search can yield unexpected winners from diverse ancestry, challenging the dominance of sequential champion approaches in program evolution.
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.
