News for hackers. Stuff that matters.
Re:AlphaEvolve: A Gemini-powered coding agent for designing ... by throwaway314155 on Wednesday May 14, 2025 @06:36PM (#43989902)
> which creates a situation where Gemini 2.0 was used in a way to train Gemini 2.5.

The use of synthetic data from prior models to create both superior models and distilled models has been going on since at least OpenAI's introduction of RLHF, and probably before that too.

Re:AlphaEvolve: A Gemini-powered coding agent for designing ... by happyopossum on Thursday May 15, 2025 @01:28AM (#43992127)
> The use of synthetic data from prior models

That’s distinct from those prior models providing actual code to improve the next model