Structural Self-Supervised Objectives for Transformers

Open paper
Benchmark Model Rank Results
question-answering-on-trecqaTANDA DeBERTa-V3-Large + ALL#1MAP: 0.954MRR: 0.984
question-answering-on-wikiqaTANDA-DeBERTa-V3-Large + ALL#1MAP: 0.927MRR: 0.939

Rank counts only results with a code link.