BrowseComp
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
BrowseComp-Plus
58Accuracy
12
BrowseComp
49Success Rate
12
Browsecomp
51.5Accuracy
12
BrowseComp-ZH (test)
21.45LJFT
12
BrowseComp+ (BC+)
61.9Recall
11
BrowseComp ZH
84.4Avg@3
11
BrowseComp
28.3Accuracy
10
BrowseComp Domains (Domain-specific (9) + Full Search)
27.8Accuracy
10
BrowseComp Domain-specific (9) Search
22.5Accuracy
10
BrowseComp
58.3Pass@1 Rate
9
BrowseComp-EN (test)
61.5Pass@1
9
BrowseComp
15.1F1
9
BrowseComp+ Subtasks 2nd Pass
98.4Accuracy
8
BrowseComp
2.7F1 (Avg)
8
BrowseComp
61.3Score
7
BrowseComp-Plus
83Score
7
BrowseComp-Plus
88.5Accuracy
7
BrowseComp-Plus
42.7Accuracy
7
BrowseComp-ZH
76.1Score
7
BrowseComp-Plus
15.3nDCG@10
6
BrowseComp-zh (BC-zh)
58Pass@3
6
BrowseComp-Plus
32.33Exact Match (EM)
5
BrowseComp-zh
81.3Pass@1
5
BrowseComp en
67.6Pass@1
5
BrowseComp
9.45Accuracy
5