researchOpenAI
Go-Browse trains 7B model to beat GPT-4o mini on web navigation tasks
Researchers propose Go-Browse, a method for training web agents through structured exploration that frames data collection as graph search. A 7B parameter language model fine-tuned on 10K trajectories achieves 21.7% success on the WebArena benchmark, outperforming GPT-4o mini by 2.4 percentage points.