SEDIMA: Cross-Run Hierarchical Insight Memory for Evolutionary Search Agents

arXiv:2610.02361v1 Announce Type: cross Abstract: Large language model (LLM)-driven evolutionary search is a powerful paradigm for automated program and algorithm discovery, yet existing systems are largely memoryless: each run explores from scratch, so agents repeatedly rediscover the same improvements and re-encounter the same dead ends. We introduce SEDIMA, a persistent hierarchical insight memory for evolutionary search agents. SEDIMA distills raw traces into natural-language insights, clusters them by semantic similarity using attention-weighted centroids, and retrieves relevant guidance to condition future mutations, accumulating transferable knowledge across runs and problems rather than within a single trajectory. As a drop-in module that leaves the search operators unmodified, SEDIMA improves average final performance by 5.5% on AlgoTune and 6.6% on ALE-Bench LITE under a fixed budget of 100 evaluated candidates. Under OpenEvolve, SEDIMA requires 32.3% fewer iterations on average to reach baseline-best performance across the five evaluated backbones.

Related

• Relay, Don't Route: Adaptive Population Handoff for Cost-Efficient LLM-Driven Evolution
• Effective Harness Engineering for Algorithm Discovery with Coding Agents
• A2DEPT: Large Language Model-Driven Automated Algorithm Design via Evolutionary Program Trees
• FunFuzz: An LLM-Powered Evolutionary Fuzzing Framework

Source: arXiv cs.CL | 2026-10-05