Tag
A paper analyzing the Parameter Golf open challenge for training language models under strict size and time constraints, finding that individual techniques rarely improve BPB by more than 1% but collectively achieved a 13.6% reduction.
An autonomous research agent built by Weco became the top contributor by volume of merged records in OpenAI's Parameter Golf competition, demonstrating effective human-agent collaboration.
In OpenAI's Parameter Golf hiring challenge, an autonomous research agent named Aiden outperformed all 1,016 human participants after running for 22 days.
OpenAI publishes a post-mortem on the Parameter Golf challenge, highlighting how AI coding agents lowered experimentation costs and changed the pace of research participation. The article details technical insights from over 2,000 submissions, including innovations in quantization, training optimization, and new modeling ideas.