EvoLattice: Persistent Internal-Population Evolution for LLM-Guided Program Discovery

This paper introduces EvoLattice, a framework that represents a population of candidate programs or agent behaviors within a single directed acyclic graph, enabling fine-grained alternative-level evaluation and LLM-guided evolution.

Details

EvoLattice is a framework that represents an entire population of candidate programs or agent behaviors within a single directed acyclic graph. Each node in the graph stores multiple persistent alternatives, and every valid path through the graph defines a distinct executable candidate. This yields a large combinatorial search space without duplicating structure. EvoLattice enables fine-grained alternative-level evaluation by scoring each alternative across all paths in which it appears, producing statistics that reveal how local design choices affect global performance. These statistics provide a dense, data-driven feedback signal for LLM-guided mutation, recombination, and pruning, while preserving successful components. Structural correctness is guaranteed by a deterministic self-repair mechanism that enforces acyclicity and dependency consistency independently of the LLM. EvoLattice naturally extends to agent evolution by interpreting alternatives as prompt fragments or sub-agent behaviors. Compared to prior LLM-guided methods, EvoLattice yields more stable evolution, greater expressivity, and stronger improvement trajectories in program synthesis and optimizer meta-learning tasks.