FOLIO
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
FOLIO
89.2Accuracy
126
FOLIO (test)
95.6Accuracy
60
FOLIO
0.61Accuracy
26
FOLIO (test)
99Syntax Correctness Rate
26
FOLIO
31.53Accuracy
24
FOLIO
1.05G(W) Error
20
FOLIO
95.3AUC
18
FOLIO to GSM8K
95.1Accuracy
18
FOLIO
84.7Pass@1 Success Rate
18
FOLIO
81Accuracy
18
FOLIO-wiki-curated (test)
98.04Accuracy
17
FOLIO
85.25Initial Score
15
FOLIO
88Accuracy (FOLIO)
14
FOLIO 203 (dev)
6.4Exclusion Rate
12
FOLIO
0.22Normalized Generalization Error Approximation
10
FOLIO full expert-curated
79.9Accuracy
8
FOLIO
0.714AOC
7
FOLIO
0.35AOC
7
FOLIO (test)
66BLEU
7
FOLIO (val)
69.12Accuracy
5
FOLIO v0.0 (val)
44.1Accuracy
4
FOLIO FOL fields (val)
68.1Accuracy
4
FOLIO n=203 (held-out)
0.72Per-Class Loss
3
FOLIO
453Optimization-phase Token Usage
3
FOLIO
69.24Pass@1
2