Efficient Long-Text Understanding with Short-Text Models
Open paper
Benchmark
Model
Rank
Results
long-range-modeling-on-scrolls
BART-large SLED
#5
Avg.: 37.99
GovRep: 57.5 / 26.3 / 27.4
SumScr: 35.2 / 8.7 / 19.4
…
Rank counts only results with a code link.