Our Approach w/o IR
Reported on 1 benchmark across 1 task
Note: results are matched by exact model name. Different papers may use the same name for different model variants.
Natural Language Processing1 result
- 50.54best: 54 (OUR APPROACH)
Reported on 1 benchmark across 1 task
Note: results are matched by exact model name. Different papers may use the same name for different model variants.