BART Base

Reported on 6 benchmarks across 2 tasks · 2 papers

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Medical4 results

Language ModellingonSCROLLS
Avg.· 2022-01-10
29.01
best: 43.51 (CoLT5 XL)
SCROLLS: Standardized CompaRison Over Long Language Sequences arXiv:2201.03533
Language ModellingonSCROLLS
CNLI· 2022-01-10
77.4
best: 88.7 (UL2 20B)
SCROLLS: Standardized CompaRison Over Long Language Sequences arXiv:2201.03533
Language ModellingonSCROLLS
Nrtv· 2022-01-10
15.4
best: 31.1 (CoLT5 XL)
SCROLLS: Standardized CompaRison Over Long Language Sequences arXiv:2201.03533
Language ModellingonSCROLLS
Qspr· 2022-01-10
26.3
best: 53.9 (CoLT5 XL)
SCROLLS: Standardized CompaRison Over Long Language Sequences arXiv:2201.03533

Natural Language Processing2 results

Code GenerationonCoNaLa-Ext
BLEU· 2021-06-08
34.35
best: 35.32 (BART W/ Mined)
Reading StackOverflow Encourages Cheating: Adding Question Text Improves Extractive Code Generation arXiv:2106.04447
Code GenerationonCoNaLa
BLEU· 2021-06-08
26.24
best: 44.32 (PanGu-Coder-FT-I)
Reading StackOverflow Encourages Cheating: Adding Question Text Improves Extractive Code Generation arXiv:2106.04447