Tevatron/browsecomp-plus
收藏Hugging Face2025-12-20 更新2025-09-13 收录
下载链接：
https://hf-mirror.com/datasets/Tevatron/browsecomp-plus
下载链接
链接失效反馈官方服务：
资源简介：
BrowseComp-Plus是一个用于隔离检索器和LLM代理效果的Deep-Research系统的新基准测试。该数据集包含来自OpenAI的BrowseComp的挑战性、推理密集型查询，并且评估针对一个固定的、经过精心策划的约10万网络文档集合。数据集包含query_id、query、answer以及证据文档、黄金文档和负样本文档的列表，每个文档都有唯一标识符、文本和URL。
BrowseComp-Plus is a new benchmark for Deep-Research system that isolates the effect of the retriever and the LLM agent to enable fair, transparent comparisons of Deep-Research agents. The dataset includes challenging, reasoning-intensive queries from OpenAIs BrowseComp, evaluated against a fixed, curated corpus of approximately 100,000 web documents. The dataset consists of query_id, query, answer, and lists of evidence documents, gold documents, and negative documents, each with a unique identifier, text, and URL.
提供机构：
Tevatron
