Using latent-agent swarms for math discovery and optimization [R]

Google DeepMind's AlphaEvolve, an LLM-driven program-evolution system running on Gemini models, found a faster method for multiplying 4x4 complex matrices, breaking a record that had stood since 1969. The approach has an LLM write a program, score it, and iteratively improve it, and the article notes similar open-source efforts are emerging. The result matters because it shows latent-agent swarms can produce new mathematical discoveries and optimization gains.

Last year Google DeepMind's AlphaEvolve found a faster way to multiply 4x4 complex matrices, beating a record that had stood since 1969. The idea is to let an LLM write a program, score it, and iteratively improve it. So it's called " evolve ". But AlphaEvolve runs on Gemini models. Similar open-sou