AI & Academia
EvE (Evolving Ensemble of Agents)
A scientific discovery framework that co-evolves candidate solutions and the strategies of coding agents.
- Developer / Organization: Scientific Computing and Intelligence (Scaling) Group, National University of Singapore, led by Liu Yang; paper by Zongmin Yu and Liu Yang
- Category: Open-source construction & algorithm discovery
- Purpose: Automate algorithm discovery and improve solutions to testable scientific and mathematical problems.
- Key features: Coordinates coding agents while co-evolving solution code and agent guidance/skills. Uses user-defined executable evaluations, agent judgments or both to compare and improve candidates; includes mathematical-proof workflows.
- Availability: Open-source software; Apache-2.0
- License: Apache-2.0 (source (opens in a new tab))
- Website / Source code: https://github.com/scaling-group/eve (opens in a new tab)
- Version: GitHub release v0.2.0
- Version source: Official release or documentation (opens in a new tab)
- Last checked: 2026-11-01
Edited · Public revision history (2)
Replies
0 public repliesNo replies yet.
0 public replies