20
BrowseComp-Plus
🥇
Fair and Transparent Evaluation of Deep-Research Agents
BrowseComp-Plus
-
Tevatron/browsecomp-plus
Viewer • Updated • 830 • 968 • 15Note BrowseComp-Plus query set, containing query, positive and hard negative documents
Tevatron/browsecomp-plus-corpus
Viewer • Updated • 100k • 2.96k • 12Note BrowseComp-Plus document set
Tevatron/browsecomp-plus-indexes
Updated • 407 • 3Note Prebuilt index for BM25 and Qwen3-Embedding-8B on BrowseComp-Plus corpus
BrowseComp-Plus: A More Fair and Transparent Evaluation Benchmark of Deep-Research Agent
Paper • 2508.06600 • Published • 39Note Original BrowseComp-Plus paper
Open Data Synthesis For Deep Research
Paper • 2509.00375 • Published • 68Note [external] Cool paper eval on BrowseComp-Plus
Scaling Long-Horizon LLM Agent via Context-Folding
Paper • 2510.11967 • Published • 3Note [external] Cool paper eval on BrowseComp-Plus