Prépublication · 2026
SwarmEvolve: Latent Agent Swarms for Evolutionary Discovery and Optimization
Qile Jiang, George Em Karniadakis — Monde
Source : arXiv
Résumé
LLM-guided evolutionary search uses AI agents to improve programs for mathematical and computational problems by iteratively refining solutions and combining high-scoring methods. However, the process is often slow because the LLM generates a long textual chain of thought for each round and must repeat this generation many times to maintain population diversity. We present SwarmEvolve, a framework in which hundreds of low-cost agents per round work in swarms, reason in latent space without decoding to text, and share information by passing KV caches. Each swarm studies the problem from a diffe…