Internet Search is Not a Naive Information Retrieval Problem

"During RL training, we employ a curriculum-based rollout strategy that
incrementally degrades the quality of generated documents, progressively eliciting the model’s reasoning ability by exposing it to increasingly challenging retrieval scenarios. Extensive experiments demonstrate that ZEROSEARCH effectively incentivizes the search capabilities of LLMs using a 3B LLM as the retrieval module. Remarkably, a 7B retrieval module achieves comparable performance to the real search engine, while a 14B retrieval module even surpasses it."
https://arxiv.org/pdf/2505.04588

Simulating Internet search performance in a non-adversarial setting misses the most important part of running a search engine: the extraordinary challenge of dealing with the 'SEO' industry. Classically, you don't want to open-source the algorithm as the race to optimize against that will start almost immediately.