AlphaEvolve: A coding agent for scientific and algorithmic discovery

A preprint presenting AlphaEvolve, a coding agent for scientific and algorithmic discovery. Both papers cite it for the point that AlphaEvolve improves only what an automatic evaluator can score. It illustrates trial-bound tempo (P6): where fitness is scored automatically, trials take minutes and lineages should evolve fastest, which prediction 2 tests.

Cited in

both papers (Flagship: Evolution by design; Companion §6.3)

Used for

P6Prediction 2P6; prediction 2

Strand

Artificial life & machine evolution · Self-improving AI & machine heredity

Cite

@misc{novikov2025alphaevolve,
  title = {AlphaEvolve: A coding agent for scientific and algorithmic discovery},
  author = {Alexander Novikov and Ngân V\ u and Marvin Eisenberger and Emilien Dupont and Po-Sen Huang and Adam Zsolt Wagner and Sergey Shirobokov and Borislav Kozlovskii and Francisco J. R. Ruiz and Abbas Mehrabian and et al.},
  year = {2025},
  url = {https://arxiv.org/abs/2506.13131},
}