SwarmEvolve: Latent Agent Swarms for Evolutionary Discovery and Optimization
cs.AI, cs.NE
Submitted: 2026-11-05
Updated: 2026-11-05
Code: https://github.com/Qile-J/SwarmEvolve
Project page: https://erich-friedman.github.io/packing/hexinhex
License: http://creativecommons.org/licenses/by-nc-nd/4.0/
The gist: LLM-guided evolutionary search uses AI agents to improve programs for mathematical and computational problems by iteratively refining solutions and combining high-scoring methods.
Terminology
Abstract
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 different angle. A novelty filter keeps the most distinct agent ideas by directly measuring the distances between the agents' hidden state vectors. Crossover of ideas happens naturally in latent space, as a summarizer agent receives the combined KV caches of all swarm members. On 21 problems in mathematical discovery and scientific computing, SwarmEvolve gets the highest average score and needs a median of only 3 to 8 minutes to reach the best scores that other state-of-the-art frameworks reach in 60 minutes. On several problems, SwarmEvolve with Qwen3 8B improves on the solutions that AlphaEvolve first found with frontier LLMs. Our work shows that agent swarms that communicate in latent space offer a fast, low-cost approach to program search for scientific problems.
Sources
- CodeEvolve: an open source evolutionary coding agent for algorithmic discovery and optimization
- Global Optimization for Combinatorial Geometry Problems Revisited in the Era of LLMs
- Harnessing the Collective Intelligence of AI Agents in the Wild for New Discoveries
- Cogentic: Multi-Agent Orchestration for Automated Proof Discovery
- AdaEvolve: Adaptive LLM Driven Zeroth-Order Optimization
- Barbarians at the Gate: How AI is Upending Systems Research
- Autonomous Mathematical Discovery in an Open-World Multi-Agent Environment
- The minimum overlap problem revisited
- AgenticSciML: Collaborative Multi-Agent Systems for Emergent Discovery in Scientific Machine Learning
- GigaEvo: An Open Source Optimization Framework Powered By LLMs And Evolution Algorithms
- ImprovEvolve: Basin-Hopping Meets LLM-Guided Evolutionary Search
- Scientific Algorithm Discovery by Augmenting AlphaEvolve with Deep Research
- EvoX: Meta-Evolution for Automated Discovery
- AlphaEvolve: A coding agent for scientific and algorithmic discovery
- SwarmResearch: Orchestrating Coding Agents for Open-Ended Discovery
- Qwen3 Technical Report
- Structured Scaling of AI Discovery Across Diverse Scientific Domains
- AlphaResearch: Accelerating New Algorithm Discovery with Language Models
- Learning to Discover at Test Time
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents