Leaderboard

Current

Live standings from the Kaggle public leaderboards

OverviewTaskTimelineRulesParticipateLeaderboardPublicationsOrganisersFAQ
0.711
Best BalAcc@10
23
Teams
613
Submissions
—
Days remaining
Metric:

Teams by BalAcc@10

Top-10 balanced accuracy on the 50-word vocab — Kaggle’s official public score

1MindLink
0.7112
2Umur Yıldız
0.7055
3Jordan Griffith
0.6694
4JovanaLab
0.6558
5MagnetoLex
0.6548
6CWBZ
0.6533
7kakuteki
0.6463
8neural2speech
0.6447
9@Arnauya
0.6428
10SENPAI
0.6296
11onlytry
0.6152
12Brain King
0.6103
13titus fisher
0.5758
14Kim
0.5038
15Baseline
0.4444
16Shuntaro Suzuki
0.3540
17MindReaders2.0
0.3322
18monte-carlo
0.2900
19KarahanYilmazer
0.2602
20Bhargav Kowshik
0.2338

Showing top 20 of 23 teams — full standings in the table below.

Best score per team — every metric

Kaggle ranks one number. We archive every submission and re-score it ourselves, so each team's best on each metric sits side by side (public split). Click a metric to rank by it.

#TeamBalAcc@10 ↓BalAcc@1 OVMI Subs
1MindLink
0.7112
0.2392
1.1265
87
2Umur Yıldız
0.7055
0.2432
1.1690
64
3Jordan Griffith
0.6694
0.2386
1.1628
28
4JovanaLab
0.6558
0.2729
1.0757
26
5MagnetoLex
0.6548
0.2173
1.1227
34
6CWBZ
0.6533
0.1266
1.0326
3
7kakuteki
0.6463
0.2239
1.1474
47
8neural2speech
0.6447
0.1447
1.1265
10
9@Arnauya
0.6428
0.1899
1.1423
71
10SENPAI
0.6296
0.1858
1.0606
41
11onlytry
0.6152
0.1934
1.0748
9
12Brain King
0.6103
0.1522
1.1299
36
13titus fisher
0.5758
0.1472
1.0546
25
14Kim
0.5038
0.0874
0.9822
52
15Baseline
0.4444
0.0733
0.9803
1
16Shuntaro Suzuki
0.3540
0.0488
0.8104
5
17MindReaders2.0
0.3322
0.0740
0.9441
49
18monte-carlo
0.2900
0.0483
0.4671
5
19KarahanYilmazer
0.2602
0.0405
0.9752
9
20Bhargav Kowshik
0.2338
0.0292
1.0947
6
21lavender_lover
0.1906
0.0200
0.4893
3
22LizaSheina
0.1880
0.0284
0.3681
1
23Viknesh .V
0.1600
0.0200
0.0000
1

About these numbers

BalAcc@10 is Kaggle's official public score, shown here exactly as Kaggle reports it. BalAcc@1 (strict top-1) and OVMI (the information a decoder recovers over the vocabulary) are computed by us from the submitted files, on the same public split — final placements use the private split, revealed after the deadline.

Updated … · Deep Track on Kaggle →